-The feature lets you specify a deletion threshold, above which an admin

-needs to explicitly choose to allow the deletions to be processed.

- If the deletion threshold is met, an email will be sent.

-When the deletion threshold is met, the job will go into quarantine and a notification email will be sent. The quarantined job can then be allowed or rejected. To learn more about quarantine behavior, see [Application provisioning in quarantine status](application-provisioning-quarantine-status.md).

-If you encounter an accidental deletion, you'll see it on the provisioning status page. It will say **Provisioning has been quarantined. See quarantine details for more information**.

-The **Allow deletes** action will delete the objects that triggered the accidental delete threshold. Use the following procedure to accept the deletes.

-3. You'll see confirmation that the deletions were accepted and the status will return to healthy with the next cycle.

-If you don't want to allow the deletions, you need to do the following:

-You can test the feature by triggering disable / deletion events by setting the threshold to a low number, for example 3, and then changing scoping filters, un-assigning users, and deleting users from the directory (see common scenarios in next section).

-Let the provisioning job run (20 ΓÇô 40 mins) and navigate back to the provisioning page. You'll see the provisioning job in quarantine and can choose to allow the deletions or review the provisioning logs to understand why the deletions occurred.

-When a user is set to be removed from the target application (or target tenant), it will be counted against the

-single cycle, the ΓÇ£circuit breakerΓÇ¥ wonΓÇÖt be triggered. If multiple cycles are needed to reach a

-steady state, the deletion threshold will be evaluated per cycle.

-|Enable Export Password in First Pass|Checked|Not supported. This will be ignored.|

-|Import Script|Paste ImportData code as value|

-|Begin Export Script|Paste Begin export code as value|

-|Export Script|Paste ExportData code as value|

- "clientConfiguration": {

- "timeoutInMilliseconds": 2000,

- "maximumRetries": 1

- },

-There are two types of authentication available for service principals: password-based authentication (application secret) and certificate-based authentication. *We recommend using a certificate*, but you can also create an application secret.

-### Option 1 (recommended): Create and upload a self-signed certificate

-You can use an existing certificate if you've one. Optionally, you can create a self-signed certificate for *testing purposes only*. To create a self-signed certificate, open Windows PowerShell and run [New-SelfSignedCertificate](/powershell/module/pki/new-selfsignedcertificate) with the following parameters to create the certificate in the user certificate store on your computer:

-After registering the certificate with your application in the application registration portal, enable the client application code to use the certificate.

-### Option 2: Create a new application secret

-When you've completed this tutorial, your application will accept sign-ins of personal Microsoft accounts (including outlook.com, live.com, and others) and work or school accounts from any company or organization that uses Azure AD.

-The app in this tutorial will sign in users and get data on their behalf. This data will be accessed through a protected API (Microsoft Graph API) that requires authorization and is protected by the Microsoft identity platform.

- implementation 'com.microsoft.identity.client:msal:4.2.0'

- SingleAccount,

- MultipleAccount,

- B2C

- if (id == R.id.nav_multiple_account) {

- setCurrentFragment(AppFragment.MultipleAccount);

- }

- if (id == R.id.nav_b2c) {

- setCurrentFragment(AppFragment.B2C);

- }

- case MultipleAccount:

- getSupportActionBar().setTitle("Multiple Account Mode");

- return;

- case B2C:

- getSupportActionBar().setTitle("B2C Mode");

- return;

- case MultipleAccount:

- attachFragment(new MultipleAccountModeFragment());

- return;

- case B2C:

- attachFragment(new B2CModeFragment());

- return;

-If you would like to model your UI off this tutorial, the following is a sample **activity_main.xml**.

- <LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"

- android:id="@+id/activity_main"

- android:background="#FFFFFF"

- android:orientation="vertical"

- <LinearLayout

- android:orientation="horizontal"

- android:paddingTop="5dp"

- android:paddingBottom="5dp"

- android:weightSum="10">

- <Button

- android:id="@+id/signIn"

- android:layout_width="0dp"

- android:layout_height="wrap_content"

- android:layout_weight="5"

- android:gravity="center"

- android:text="Sign In"/>

- <Button

- android:id="@+id/clearCache"

- android:layout_width="0dp"

- android:layout_height="wrap_content"

- android:layout_weight="5"

- android:gravity="center"

- android:text="Sign Out"

- android:enabled="false"/>

- </LinearLayout>

- <LinearLayout

- android:layout_width="match_parent"

- android:layout_height="wrap_content"

- android:gravity="center"

- android:orientation="horizontal">

- <Button

- android:id="@+id/callGraphInteractive"

- android:layout_width="0dp"

- android:layout_height="wrap_content"

- android:layout_weight="5"

- android:text="Get Graph Data Interactively"

- android:enabled="false"/>

- <Button

- android:id="@+id/callGraphSilent"

- android:layout_width="0dp"

- android:layout_height="wrap_content"

- android:layout_weight="5"

- android:text="Get Graph Data Silently"

- android:enabled="false"/>

- </LinearLayout>

- android:text="Getting Graph Data..."

- android:textColor="#3f3f3f"

- android:layout_height="wrap_content"

- android:layout_marginLeft="5dp"

- android:id="@+id/graphData"

- android:visibility="invisible"/>

- android:id="@+id/current_user"

- android:layout_height="0dp"

- android:layout_marginTop="20dp"

- android:layout_weight="0.8"

- android:text="Account info goes here..." />

- android:id="@+id/txt_log"

- android:layout_height="0dp"

- android:layout_marginTop="20dp"

- android:layout_weight="0.8"

- android:text="Output goes here..." />

-[](media/multilateral-federation-solution-one/azure-ad-cirrus-bridge.png#lightbox)

-* **External attribute store support** - [Azure AD custom claims provider](../develop/custom-claims-provider-overview.md) enables you to use an external attribute store (like an external LDAP Directory) to add additional claims into tokens on a per app basis. It uses a custom extension that calls an external REST API to fetch claims from external systems.

-1. On the email customization screen, enter a custom subject, message body, and the email language translation option that will be used to translate the message body of the email.

-$accesspackage = Get-MgEntitlementManagementAccessPackage -DisplayNameEq "Marketing Campaign" -ExpandProperty "accessPackageAssignmentPolicies"

-$policy = $accesspackage.AccessPackageAssignmentPolicies[0]

-$accesspackage = Get-MgEntitlementManagementAccessPackage -DisplayNameEq "Marketing Campaign" -ExpandProperty "accessPackageAssignmentPolicies"

-$policy = $accesspackage.AccessPackageAssignmentPolicies[0]

-$accesspackage = Get-MgEntitlementManagementAccessPackage -DisplayNameEq "Marketing Campaign" -ExpandProperty "accessPackageAssignmentPolicies"

-$policy = $accesspackage.AccessPackageAssignmentPolicies[0]

-## Configure Conditional Access policy

-## Add protected actions

-## Test protected actions

- Configure a Conditional Access authentication context and an associated Conditional Access policy. Protected actions use an authentication context, which allows policy enforcement for fine-grain resources in a service, like Azure AD permissions. A good policy to start with is to require passwordless MFA and exclude an emergency account. [Learn more](./protected-actions-add.md#configure-conditional-access-policy)

- Add protected actions by assigning Conditional Access authentication context values to selected permissions. [Learn more](./protected-actions-add.md#add-protected-actions)

- Sign in as a user and test the user experience by performing the protected action. You should be prompted to satisfy the Conditional Access policy requirements. For example, if the policy requires multi-factor authentication, you should be redirected to the sign-in page and prompted for strong authentication. [Learn more](./protected-actions-add.md#test-protected-actions)

-* Asana supports **SP** initiated SSO

-* Asana supports [**automated** user provisioning](asana-provisioning-tutorial.md)

-1. In a different browser window, sign on to your Asana application. To configure SSO in Asana, access the workspace settings by clicking the workspace name on the top right corner of the screen. Then, click on **\<your workspace name\> Settings**.

- 

-2. On the **Organization settings** window, click **Administration**. Then, click **Members must log in via SAML** to enable the SSO configuration. The perform the following steps:

- 

- a. In the **Sign-in page URL** textbox, paste the **Login URL**.

- b. Right click the certificate downloaded from Azure portal, then open the certificate file using Notepad or your preferred text editor. Copy the content between the begin and the end certificate title and paste it in the **X.509 Certificate** textbox.

-3. Click **Save**. Go to [Asana guide for setting up SSO](https://asana.com/guide/help/premium/authentication#gl-saml) if you need further assistance.

- 

-In this tutorial, you'll learn how to integrate Hootsuite with Azure Active Directory (Azure AD). When you integrate Hootsuite with Azure AD, you can:

-1. On the **Basic SAML Configuration** section, if you wish to configure the application in **IDP** initiated mode, perform the following step:

- In the **Reply URL** text box, type a URL using one of the following patterns:

- | Reply URL |

- ||

- |`https://hootsuite.com/member/sso-complete`|

- |`https://hootsuite.com/sso/<ORG_ID>`|

- |

-1. Click **Set additional URLs** and perform the following step if you wish to configure the application in **SP** initiated mode:

- > [!NOTE]

- > These values are not real. Update these values with the actual Reply URL. Contact [Hootsuite Client support team](https://hootsuite.com/about/contact-us#) to get these values. You can also refer to the patterns shown in the **Basic SAML Configuration** section in the Azure portal.

-In this section, you'll create a test user in the Azure portal called B.Simon.

-In this section, you'll enable B.Simon to use Azure single sign-on by granting access to Hootsuite.

-# Tutorial: Azure Active Directory integration with Marketo

-* An Azure AD subscription. If you don't have an Azure AD environment, you can get one-month trial [here](https://azure.microsoft.com/pricing/free-trial/)

-* Marketo single sign-on enabled subscription

-* Marketo supports **IDP** initiated SSO

-## Adding Marketo from the gallery

- Alternatively, you can also use the [Enterprise App Configuration Wizard](https://portal.office.com/AdminPortal/home?Q=Docs#/azureadappintegration). In this wizard, you can add an application to your tenant, add users/groups to the app, assign roles, as well as walk through the SSO configuration as well. [Learn more about Microsoft 365 wizards.](/microsoft-365/admin/misc/azure-ad-setup-guides)

-### Configure Azure AD SSO

-1. On the **Basic SAML Configuration** section, enter the values for the following fields:

-5. On the **Set up Single Sign-On with SAML** page, in the **SAML Signing Certificate** section, click **Download** to download the **Certificate (Base64)** from the given options as per your requirement and save it on your computer.

-6. On the **Set up Marketo** section, copy the appropriate URL(s) as per your requirement.

-In this section, you'll create a test user in the Azure portal called B.Simon.

-In this section, you'll enable B.Simon to use Azure single sign-on by granting access to Marketo.

-3. Navigate to the **Security** menu and click **Users & Roles**

-4. Click the **Invite New User** link on the Users tab

-5. In the Invite New User wizard fill the following information

- b. Enter the **First Name** in the textbox

- c. Enter the **Last Name** in the textbox

- d. Click **Next**

-6. In the **Permissions** tab, select the **userRoles** and click **Next**

-### Test SSO

-# Tutorial: Azure Active Directory single sign-on (SSO) integration with SAP Cloud for Customer

-In this tutorial, you'll learn how to integrate SAP Cloud for Customer with Azure Active Directory (Azure AD). When you integrate SAP Cloud for Customer with Azure AD, you can:

-* SAP Cloud for Customer supports **SP** initiated SSO

-## Adding SAP Cloud for Customer from the gallery

- Alternatively, you can also use the [Enterprise App Configuration Wizard](https://portal.office.com/AdminPortal/home?Q=Docs#/azureadappintegration). In this wizard, you can add an application to your tenant, add users/groups to the app, assign roles, as well as walk through the SSO configuration as well. [Learn more about Microsoft 365 wizards.](/microsoft-365/admin/misc/azure-ad-setup-guides)

-In this section, you'll create a test user in the Azure portal called B.Simon.

-In this section, you'll enable B.Simon to use Azure single sign-on by granting access to SAP Cloud for Customer.

-2. From the left side of menu, click on **Identity Providers** > **Corporate Identity Providers** > **Add** and on the pop-up add the Identity provider name like **Azure AD**, click **Save** then click on **SAML 2.0 Configuration**.

- 

- b. Once the XML file is successfully uploaded, the below values will get auto populated automatically then click **Save**.

-Once you configure the SAP Cloud for Customer you can enforce session controls, which protects exfiltration and infiltration of your organizationΓÇÖs sensitive data in real time. Session controls extends from Conditional Access. [Learn how to enforce session control with Microsoft Defender for Cloud Apps](/cloud-app-security/proxy-deployment-aad).

-> Prior to Windows OS Build 20348.1668, there was a limitation around Windows Overlay pods incorrectly SNATing packets from host network pods, this had a more detrimental effect for clusters upgrading to Overlay. To avoid this issue, **use Windows OS Build 20348.1668**

-This network disruption will only occur during the upgrade. Once the migration to Overlay has completed for all node pools, all Overlay pods will be able to communicate successfully with the Windows pods.

-A persistent volume represents a piece of storage that has been provisioned for use with Kubernetes pods. A persistent volume can be used by one or many pods, and can be dynamically or statically provisioned. This article shows you how to dynamically create persistent volumes with Azure Disks for use by a single pod in an Azure Kubernetes Service (AKS) cluster.

-> An Azure disk can only be mounted with *Access mode* type *ReadWriteOnce*, which makes it available to one pod in AKS. If you need to share a persistent volume across multiple pods, use [Azure Files][azure-files-pvc].

-> * The cluster autoscaler profile affects **all node pools** that use the cluster autoscaler. You can't set an autoscaler profile per node pool. When you est the profile, any existing node pools with the cluster autoscaler enabled immediately start using the profile.

-* [Azure Private Link service limitations][private-link-service] apply to private clusters.

-You use the managed identity in the next steps to grant permissions to manage the Azure DNS zone and retrieve certificates from the Azure Key Vault.

- ```azurecli-interactive

- az keyvault set-policy --name <KeyVaultName> --object-id $MANAGEDIDENTITY_OBJECTID --secret-permissions get --certificate-permissions get

- ```

-| .NET | [Azure.Identity](https://learn.microsoft.com/dotnet/api/overview/azure/identity-readme) | 1.9.0-beta.2 | [Link](https://github.com/Azure/azure-workload-identity/tree/main/examples/azure-identity/dotnet) |

-| Go | [azidentity](https://pkg.go.dev/github.com/Azure/azure-sdk-for-go/sdk/azidentity) | 1.3.0-beta.1 | [Link](https://github.com/Azure/azure-workload-identity/tree/main/examples/azure-identity/go) |

-| Java | [azure-identity](https://learn.microsoft.com/java/api/overview/azure/identity-readme) | 1.9.0-beta.1 | [Link](https://github.com/Azure/azure-workload-identity/tree/main/examples/azure-identity/java) |

-| JavaScript | [@azure/identity](https://learn.microsoft.com/javascript/api/overview/azure/identity-readme) | 3.2.0-beta.1 | [Link](https://github.com/Azure/azure-workload-identity/tree/main/examples/azure-identity/node) |

-| Python | [azure-identity](https://learn.microsoft.com/python/api/overview/azure/identity-readme) | 1.13.0b2 | [Link](https://github.com/Azure/azure-workload-identity/tree/main/examples/azure-identity/python) |

-* Learn more about [upcoming breaking changes](breaking-changes/overview.md) in API Management.

-|**5**|**Learn more**|[Frequently asked questions](migrate.md#frequently-asked-questions)<br><br>[Community support](https://aka.ms/asev1v2retirement)|

-> [Learn about App Service Environment v3](overview.md)

-App Service offers automatic scaling that adjusts the number of instances based on incoming HTTP requests. Automatic scaling guarantees that your web apps can manage different levels of traffic. You can adjust scaling settings, like setting the minimum and maximum number of instances per web app, to enhance performance. The platform tackles cold start issues by prewarming instances that act as a buffer when scaling out, resulting in smooth performance transitions. Billing is calculated per second using existing meters, and prewarmed instances are also charged per second.

-It's common to deploy multiple web apps to a single App Service Plan. You can enable automatic scaling for an App Service Plan and configure a range of instances for each of the web apps. As your web app starts receiving incoming HTTP traffic, App Service monitors the load and adds instances. Resources may be shared when multiple web apps within an App Service Plan are required to scale out simultaneously.

-> [!IMPORTANT]

-> [`Always ON`](./configure-common.md?tabs=portal#configure-general-settings) needs to be disabled to use automatic scaling.

->

-To enable automatic scaling in the Azure portal, select **Scale out (App Service Plan)** in the web app's left menu. Select **Automatic (preview)**, update the __Maximum burst__ value, and select the **Save** button.

-To set the minimum number of instances in the Azure portal, select **Scale out (App Service Plan)** in the web app's left menu, update the **Always ready instances** value, and select the **Save** button.

-To set the maximum number of web app instances in the Azure portal, select **Scale out (App Service Plan)** in the web app's left menu, select **Enforce scale out limit**, update the **Maximum scale limit**, and select the **Save** button.

-To disable automatic scaling in the Azure portal, select **Scale out (App Service Plan)** in the web app's left menu, select **Manual**, and select the **Save** button.

-## Frequently asked questions

-### How is automatic scaling different than autoscale?

-Automatic scaling is a new scaling option in App Service that automatically handles web app scaling decisions for you. **[Azure autoscale](../azure-monitor/autoscale/autoscale-overview.md)** is a pre-existing Azure capability for defining schedule-based and resource-based scaling rules for your App Service Plans.

-A comparison of scale out and scale in options available on App Service:

-| | **Manual scaling** | **Auto scaling** | **Automatic scaling** |

-| | | | |

-| Available pricing tiers | Basic and Up | Standard and Up | Premium v2 and Premium v3 |

-|Rule-based scaling |No |Yes |No, the platform manages the scale out and in based on HTTP traffic. |

-|Schedule-based scaling |No |Yes |No|

-|Always ready instances | No, your web app runs on the number of manually scaled instances. | No, your web app runs on other instances available during the scale out operation, based on threshold defined for autoscale rules. | Yes (minimum 1) |

-|Prewarmed instances |No |No |Yes (default 1) |

-|Per-app maximum |No |No |Yes|

-### How does automatic scaling work with existing autoscale rules?

-Once automatic scaling is configured, existing Azure autoscale rules and schedules are disabled. Applications can use either automatic scaling, or autoscale, but not both.

-### Does automatic scaling support Azure Function apps?

-### How to monitor the current instance count and instance history?

-Use Application Insights [Live Metrics](../azure-monitor/app/live-stream.md) to check the current instance count, and [performanceCounters](../azure-functions/analyze-telemetry-data.md#query-telemetry-data) to check the instance count history.

-> You can create private and public listeners with the same port number (Preview feature). However, be aware of any Network Security Group (NSG) associated with the application gateway subnet. Depending on your NSG's configuration, you may need an inbound rule with **Destination IP addresses** as your application gateway's public and private frontend IPs.

-> - Destination IP addresses: Public and Private frontend IPs of your application gateway.

-description: Form Recognizer software development kits (SDKs) expose Form Recognizer models, features and capabilities, using C#, Java, JavaScript, or Python programming language.

-| [.NET/C# → 4.0.0 (latest GA release)](https://azuresdkdocs.blob.core.windows.net/$web/dotnet/Azure.AI.FormRecognizer/4.0.0/https://docsupdatetracker.net/index.html)|[NuGet](https://www.nuget.org/packages/Azure.AI.FormRecognizer)|[**v3.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-2022-08-31/operations/AnalyzeDocument)</br> [**v2.1**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2-1/operations/AnalyzeBusinessCardAsync)</br>[**v2.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2/operations/AnalyzeLayoutAsync) |[Windows, macOS, Linux, Docker](https://dotnet.microsoft.com/download)|

-|[Java → 4.0.6 (latest GA release)](https://azuresdkdocs.blob.core.windows.net/$web/java/azure-ai-formrecognizer/4.0.0/https://docsupdatetracker.net/index.html) |[MVN repository](https://mvnrepository.com/artifact/com.azure/azure-ai-formrecognizer/4.0.0-beta.6) |[**v3.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-2022-08-31/operations/AnalyzeDocument)</br> [**v2.1**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2-1/operations/AnalyzeBusinessCardAsync)</br>[**v2.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2/operations/AnalyzeLayoutAsync) |[Windows, macOS, Linux](/java/openjdk/install)|

-|[JavaScript → 4.0.0 (latest GA release)](https://azuresdkdocs.blob.core.windows.net/$web/javascript/azure-ai-form-recognizer/4.0.0/https://docsupdatetracker.net/index.html)| [npm](https://www.npmjs.com/package/@azure/ai-form-recognizer)| [**v3.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-2022-08-31/operations/AnalyzeDocument)</br> [**v2.1**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2-1/operations/AnalyzeBusinessCardAsync)</br>[**v2.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2/operations/AnalyzeLayoutAsync) | [Browser, Windows, macOS, Linux](https://nodejs.org/en/download/) |

-|[Python → 3.2.0 (latest GA release)](https://azuresdkdocs.blob.core.windows.net/$web/python/azure-ai-formrecognizer/3.2.0/https://docsupdatetracker.net/index.html) | [PyPI](https://pypi.org/project/azure-ai-formrecognizer/3.2.0/)| [**v3.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-2022-08-31/operations/AnalyzeDocument)</br> [**v2.1**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2-1/operations/AnalyzeBusinessCardAsync)</br>[**v2.0**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-v2/operations/AnalyzeLayoutAsync) |[Windows, macOS, Linux](/azure/developer/python/configure-local-development-environment?tabs=windows%2Capt%2Ccmd#use-the-azure-cli)

-|**.NET/C#**</br> **Java**</br> **JavaScript**</br>| 4.0.0 (latest GA release)| v3.0 / 2022-08-31 (default)| **DocumentAnalysisClient**<br>**DocumentModelAdministrationClient** |

-|**.NET/C#**</br> **Java**</br> **JavaScript**</br>| 3.1.x | v2.1 (default)</br>v2.0 | **FormRecognizerClient**</br>**FormTrainingClient** |

-|**.NET/C#**</br> **Java**</br> **JavaScript**</br>| 3.0.x| v2.0 | **FormRecognizerClient**</br>**FormTrainingClient** |

-| **Python**| 3.2.x (latest GA release) | v3.0 / 2022-08-31 (default)| **DocumentAnalysisClient**</br>**DocumentModelAdministrationClient**|

-| **Python** | 3.1.x | v2.1 (default)</br>v2.0 |**FormRecognizerClient**</br>**FormTrainingClient** |

-| **Python** | 3.0.0 | v2.0 |**FormRecognizerClient**</br>**FormTrainingClient** |

-description: Learn how to convert to the new App Configuration Spring Boot Library from the previous version.

-# Convert to new App Configuration Spring Boot library

-A new version of the App Configuration library for Spring Boot is now available. The version introduces new features such as Azure Spring global properties, but also some breaking changes. These changes aren't backwards compatible with configuration setups that were using the previous library version. For the following topics:

-* Group and Artifact Ids

-* Package path renamed

-* Classes renamed

-* Feature flag loading

-* Possible conflicts with Azure Spring global properties

-this article provides a reference on the change and actions needed to migrate to the new library version.

-## Group and Artifact ID changed

-All of the Azure Spring Boot libraries have had their Group and Artifact IDs updated to match a new format. The new package names are:

-> [!NOTE]

-> The 4.7.0 version is the first 4.x version of the library. This is to match the version of the other Spring Cloud Azure libraries.

-As of the 4.7.0 version, the App Configuration and Feature Management libraries are now part of the spring-cloud-azure-dependencies BOM. The BOM file makes it so that you no longer need to specify the version of the libraries in your project. The BOM automatically manages the version of the libraries.

-## Package path renamed

-The package path for the `spring-cloud-azure-feature-managment` and `spring-cloud-azure-feature-management-web` libraries have been renamed from `com.azure.spring.cloud.feature.manager` to `com.azure.spring.cloud.feature.management` and `com.azure.spring.cloud.feature.management.web`.

-* `ConfigurationClientBuilderSetup` has been renamed to `ConfigurationClientCustomizer` and its `setup` method has been renamed to `customize`

-* `SecretClientBuilderSetup` has been renamed to `SecretClientCustomizer` and its `setup` method has been renamed to `customize`

-* `AppConfigurationCredentialProvider` and `KeyVaultCredentialProvider` have been removed. Instead you can use [Azure Spring common configuration properties](/azure/developer/java/spring-framework/configuration) or modify the credentials using `ConfigurationClientCustomizer`/`SecretClientCustomizer`.

-Feature flags now support loading using multiple key/label filters.

-> [!NOTE]

-> The property `spring.cloud.azure.appconfiguration.stores[0].feature-flags.label` has been removed. Instead you can use `spring.cloud.azure.appconfiguration.stores[0].feature-flags.selects[0].label-filter` to specify a label filter.

-## Possible conflicts with Azure Spring global properties

-[Azure Spring common configuration properties](/azure/developer/java/spring-framework/configuration) enables you to customize your connections to Azure services. The new App Configuration library will picks up any global or app configuration setting configured with Azure Spring common configuration properties. Your connection to app configuration will change if the configurations have been set for another Azure Spring library.

-> [!NOTE]

-> You can override this by using `ConfigurationClientCustomizer`/`SecretClientCustomizer` to modify the clients.

-> You may now run into an issue where more than one connection method is provided as Azure Spring global properties will automatically pick up credentials, such as Environment Variables, and use them to connect to Azure services. This can cause issues if you are using a different connection method, such as Managed Identity, and the global properties are overriding it.

- * Assign [Azure Policy guest configurations](../../governance/machine-configuration/overview.md) to audit settings inside the machine. To understand the cost of using Azure Policy Guest Configuration policies with Arc-enabled servers, see Azure Policy [pricing guide](https://azure.microsoft.com/pricing/details/azure-policy/).

-* Windows Server 2012 R2, 2016, 2019, and 2022

-* Windows PowerShell 4.0 or later (already included with Windows Server 2012 R2 and later).

-## Azure Cache for Redis now supports upto 30 shards (preview)

-**Limitations**

-* Shard limit for caches with Redis verion 4 is 10.

-* Shard limit for [caches affected by cloud service retirement](./cache-faq.yml#caches-with-a-dependency-on-cloud-services--classic) is 10.

-* Maintenance will take longer as each node take roughly 20 minutes to update. Other maintenance operations will be blocked while your cache is under maintenance.

-No,scaling out using clustering and increasing the number of shards doesn't help in increasing the number of supported client connections.

-### Support for upto 30 shards for clustered Azure Cache for Redis instances

-Azure Cache for Redis now supports clustered caches with upto 30 shards which means your applications can store more data and scale better with your workloads.

-For more information, see [Configure clustering for Azure Cache for Redis instance](cache-how-to-premium-clustering.md#azure-cache-for-redis-now-supports-upto-30-shards-preview).

-The Enterprise and Enterprise Flash tiers now support the ability to scale cache instances up and out without requiring downtime or data loss. Scale up and scale out actions can both occur in the same operation.

-For more information, see [Scale an Azure Cache for Redis instance](cache-how-to-scale.md)

-Cache instances using active geo-replication now support the RedisJSON module.

-For more information, see [Configure active geo-replication](cache-how-to-active-geo-replication.md).

-Caches using active geo-replication now include a built-in _flush_ operation that can be initiated at the control plane level. Use the _flush_ operation with your cache instead of the `FLUSH ALL` and `FLUSH DB` operations, which are blocked by design for active geo-replicated caches.

-For more information, see [Flush operation](cache-how-to-active-geo-replication.md#flush-operation)

-For more information, see [Enable disk encryption](cache-how-to-encryption.md)

-For more information, see [Enabling connection audit logs](cache-monitor-diagnostic-settings.md)

-### Support for managed identity in Azure Cache for Redis

-You might have a limited number of runbook actions per action group.

-You might have a limited number of Azure app actions per action group.

-You might have a limited number of function actions per action group.

-You might have a limited number of ITSM actions per action group.

-You might have a limited number of Logic Apps actions per action group.

-You might have a limited number of SMS actions per action group.

-You might have a limited number of voice actions per action group.

-description: Overview of auto-instrumentation for Azure Monitor Application Insights codeless application performance management.

-# What is auto-instrumentation for Azure Monitor Application Insights?

-Auto-instrumentation quickly and easily enables [Application Insights](app-insights-overview.md) to make [telemetry](data-model-complete.md) like metrics, requests, and dependencies available in your [Application Insights resource](create-workspace-resource.md).

-The following table shows the current state of auto-instrumentation availability.

-|Azure App Service on Windows - Publish as Code | [ :white_check_mark: :link: ](azure-web-apps-net.md) ^[1](#OnBD) | [ :white_check_mark: :link: ](azure-web-apps-net-core.md) ^[1](#OnBD) | [ :white_check_mark: :link: ](azure-web-apps-java.md) ^[1](#OnBD) | [ :white_check_mark: :link: ](azure-web-apps-nodejs.md) ^[1](#OnBD) | :x: |

-|Azure App Service on Linux - Publish as Code | :x: | [ :white_check_mark: :link: ](azure-web-apps-net-core.md?tabs=linux) ^[1](#OnBD) | [ :white_check_mark: :link: ](azure-web-apps-java.md) ^[1](#OnBD) | [ :white_check_mark: :link: ](azure-web-apps-nodejs.md?tabs=linux) | :x: |

-|Azure App Service on Linux - Publish as Docker | :x: | :x: | [ :white_check_mark: :link: ](azure-web-apps-java.md) | [ :white_check_mark: :link: ](azure-web-apps-nodejs.md?tabs=linux) | :x: |

-|Azure Kubernetes Service (AKS) | :x: | :x: | [ :white_check_mark: :link: ](opentelemetry-enable.md?tabs=java) | :x: | :x: |

-|Azure VMs Windows | [ :white_check_mark: :link: ](azure-vm-vmss-apps.md) ^{[2](#Preview)} ^[3](#Agent) | [ :white_check_mark: :link: ](azure-vm-vmss-apps.md) ^{[2](#Preview)} ^[3](#Agent) | [ :white_check_mark: :link: ](opentelemetry-enable.md?tabs=java) | :x: | :x: |

-|On-premises VMs Windows | [ :white_check_mark: :link: ](application-insights-asp-net-agent.md) ^[3](#Agent) | [ :white_check_mark: :link: ](application-insights-asp-net-agent.md) ^{[2](#Preview)} ^[3](#Agent) | [ :white_check_mark: :link: ](opentelemetry-enable.md?tabs=java) | :x: | :x: |

-|Standalone agent - any environment | :x: | :x: | [ :white_check_mark: :link: ](opentelemetry-enable.md?tabs=java) | :x: | :x: |

-* [Application map](app-map.md)

-When navigating to Live Metrics, you may see a banner with the status message: "Data is temporarily inaccessible. The updates on our status are posted here https://aka.ms/aistatus"

-Verify if any firewalls or browser extensions are blocking access to Live Metrics. For example, some popular ad-blocker extensions block connections to \*.monitor.azure.com. In order to use the full capabilities of Live Metrics, either disable the ad-blocker extension or add an exclusion rule for the domain \*.livediagnostics.monitor.azure.com to your ad-blocker, firewall, etc.

-Heavier traffic is expected while the LiveMetrics pane is open. Navigate away from the LiveMetrics pane to restore normal traffic flow of traffic.

-Application Insights SDKs poll QuickPulse endpoints with REST API calls once every five seconds to check if the LiveMetrics pane is being viewed.

-### [.NET](#tab/net)

-// Done setting role name and role instance

-// Set ResourceBuilder on the provider.

- .AddSource("OTel.AzureMonitor.Demo")

- .AddAzureMonitorTraceExporter(o =>

- o.ConnectionString = "<Your Connection String>";

- })

- .Build();

- .AddSource("OTel.AzureMonitor.Demo")

- .SetSampler(new ApplicationInsightsSampler(0.1F))

- .AddAzureMonitorTraceExporter(o =>

- {

- o.ConnectionString = "<Your Connection String>";

- })

- .Build();

-#### [.NET](#tab/net)

-For example:

- .AddAzureMonitorTraceExporter(o => {

- o.ConnectionString = "<Your Connection String>";

- o.StorageDirectory = "C:\\SomeDirectory";

- })

- .Build();

-You might want to enable the OpenTelemetry Protocol (OTLP) Exporter alongside your Azure Monitor Exporter to send your telemetry to two locations.

-1. Install the [OpenTelemetry.Exporter.OpenTelemetryProtocol](https://www.nuget.org/packages/OpenTelemetry.Exporter.OpenTelemetryProtocol/) package along with [Azure.Monitor.OpenTelemetry.Exporter](https://www.nuget.org/packages/Azure.Monitor.OpenTelemetry.Exporter) in your project.

- // Sends data to Application Insights as well as OTLP

- using var tracerProvider = Sdk.CreateTracerProviderBuilder()

- .AddSource("OTel.AzureMonitor.Demo")

- .AddAzureMonitorTraceExporter(o =>

- {

- o.ConnectionString = "<Your Connection String>"

- })

- .AddOtlpExporter()

- .Build();

-Add `UseAzureMonitor()` to your application startup. Depending on your version of .NET Core, this will be in either your `startup.cs` or `program.cs` class.

-The following example demonstrates how the the [Runtime Instrumentation](https://www.nuget.org/packages/OpenTelemetry.Instrumentation.Runtime) can be added to collect additional metrics.

-It isn't available in .NET.

- - [HttpClient and HttpWebRequest](https://github.com/open-telemetry/opentelemetry-dotnet/blob/1.0.0-rc9.14/src/OpenTelemetry.Instrumentation.Http/README.md#filter)

-<!-- PR for Hector-->

-| cadvisor | bool | `true` | Linux daemosnet | Scrape cadvisor in every node in the K8s cluster without any extra scrape config.<br>Linux only. |

-|[Application instrumentation](app/app-insights-overview.md)| Application Insights is enabled through either [Auto-Instrumentation (agent)](app/codeless-overview.md#what-is-auto-instrumentation-for-azure-monitor-application-insights) or by adding the Application Insights SDK to your application code. For more information, reference [How do I instrument an application?](app/app-insights-overview.md#how-do-i-instrument-an-application).|

-1. In your Application Insights resource, select **Search**.

-* You cannot change the service level for volumes in a cross-region replication relationship.

-resource sa 'Microsoft.Storage/storageAccounts@2019-06-01' = if (newOrExisting == 'new') {

- tier: 'Standard'

- properties: {

- accessTier: 'Hot'

- }

-When the parameter `newOrExisting` is set to **new**, the condition evaluates to true. The storage account is deployed. However, when `newOrExisting` is set to **existing**, the condition evaluates to false and the storage account isn't deployed.

-resource vmName_omsOnboarding 'Microsoft.Compute/virtualMachines/extensions@2017-03-30' = if (!empty(logAnalytics)) {

- workspaceId: ((!empty(logAnalytics)) ? reference(logAnalytics, '2015-11-01-preview').customerId : null)

- workspaceKey: ((!empty(logAnalytics)) ? listKeys(logAnalytics, '2015-11-01-preview').primarySharedKey : null)

- >[!IMPORTANT]

- > Deployment script uses the available CLI images from Microsoft Container Registry (MCR). It takes about one month to certify a CLI image for deployment script. Don't use the CLI versions that were released within 30 days. To find the release dates for the images, see [Azure CLI release notes](/cli/azure/release-notes-azure-cli). If an unsupported version is used, the error message lists the supported versions.

-> See [Cloud Services (extended support) deployment move guidance](./move-limitations/classic-model-move-limitations.md). Cloud Services (extended support) deployment resources can be moved across subscriptions with an operation specific to that scenario.

- "functions": [],

- "resources": [

- {

- "apiVersion": "2019-06-01",

- "name": "Standard_LRS",

- "tier": "Standard"

- "kind": "StorageV2",

- "properties": {

- "accessTier": "Hot"

- }

- ]

-When the parameter `newOrExisting` is set to **new**, the condition evaluates to true. The storage account is deployed. However, when `newOrExisting` is set to **existing**, the condition evaluates to false and the storage account isn't deployed.

- "defaultValue": "mcr.microsoft.com/azuredeploymentscripts-powershell:az5.2",

-The default container image specified in the template is **mcr.microsoft.com/azuredeploymentscripts-powershell:az5.2**. See a list of all [supported Azure PowerShell versions](https://mcr.microsoft.com/v2/azuredeploymentscripts-powershell/tags/list).

- >[!IMPORTANT]

- > Deployment script uses the available CLI images from Microsoft Container Registry (MCR). It takes about one month to certify a CLI image for deployment script. Don't use the CLI versions that were released within 30 days. To find the release dates for the images, see [Azure CLI release notes](/cli/azure/release-notes-azure-cli). If an unsupported version is used, the error message lists the supported versions.

-description: In this tutorial, you'll use Azure SQL Edge for wake-detection on the turbines in a Contoso wind farm.

-# Using Azure SQL Edge to build smarter renewable resources

-This Azure SQL Edge demo is based on a Contoso Renewable Energy, a wind turbine farm that uses SQL DB edge for data processing onboard the generator.

-This demo will walk you through resolving an alert being raised because of wind turbulence being detected at the device. You will train a model and deploy it to SQL DB Edge that will correct the detected wind wake and ultimately optimize power output.

-## Setting up the demo on your local computer

-Git will be used to copy all files from the demo to your local computer.

-1. Install git from [here](https://git-scm.com/download).

-2. Open a command prompt and navigate to a folder where the repo should be downloaded.

-3. Issue the command https://github.com/microsoft/sql-server-samples.git.

-4. Navigate to **'sql-server-samples\samples\demos\azure-sql-edge-demos\Wind Turbine Demo'** in the location where the repository is cloned.

-5. Follow the instructions in README.md to set up the demo environment and execute the demo.

-Azure VMware Solution provides you with private clouds that contain VMware vSphere clusters built from dedicated bare-metal Azure infrastructure. Azure VMware Solution is available in Azure Commercial and in Public Preview in Azure Government.The minimum initial deployment is three hosts, but more hosts can be added, up to a maximum of 16 hosts per cluster. All provisioned private clouds have VMware vCenter Server, VMware vSAN, VMware vSphere, and VMware NSX-T Data Center. As a result, you can migrate workloads from your on-premises environments, deploy new virtual machines (VMs), and consume Azure services from your private clouds. For information about the SLA, see the [Azure service-level agreements](https://azure.microsoft.com/support/legal/sla/azure-vmware/v1_1/) page.

-<!-- LINKS - internal -->

-[Updating VMware HCX](https://docs.vmware.com/en/VMware-HCX/4.5/hcx-user-guide/GUID-508A94B2-19F6-47C7-9C0D-2C89A00316B9.html)

-| **Regions** | **Americas** ΓÇô Central US, East US 2, East US, North Central US, South Central US, West US 2, West US 3, West Central US, West US, Canada Central, Canada East, Brazil South <br> **Asia Pacific** ΓÇô Australia Central, Australia Central 2, Australia East, Australia Southeast, Japan East, Japan West, Korea Central, Korea South, East Asia, Southeast Asia, Central India, South India, West India, China East, China East 2, China East 3, China North, China North 2, China North 3 <br> **Europe** ΓÇô West Europe, North Europe, France Central, UK South, UK West, Germany North, Germany West Central, Switzerland North, Switzerland West, Central Switzerland North, Norway East, Norway West <br> **Africa / ME** - South Africa North, South Africa West, UAE North, UAE Central <BR> **Azure Government regions** | France South, Germany Central, Germany Northeast, US Gov IOWA |

- > * In order for the certificate to be automatically rotated to the latest version when a newer version of the certificate is available in your Key Vault, please set the certificate/secret version to 'Latest'. If a specific version is selected, you have to re-select the new version manually for certificate rotation. It takes up to 72 hours for the new version of the certificate/secret to be deployed.

-For more information, see [How to bring VMs up-to-date with the latest scale set model](../virtual-machine-scale-sets/virtual-machine-scale-sets-upgrade-scale-set.md#how-to-bring-vms-up-to-date-with-the-latest-scale-set-model)

-# Common errors and known issues when migration to Azure Cloud Services (extended support)

-| After successful prepare, linking a new Cloud Services (extended support) deployment as swappable is not allowed. | Do not link a new cloud service as swappable to a prepared cloud service. |

-| Deployment deployment-name in cloud service cloud-service-name must be within a virtual network to be migrated. | Deployment is not located in a virtual network. Refer [this](in-place-migration-technical-details.md#migration-of-deployments-not-in-a-virtual-network) document for more details. |

-| Migration of deployment deployment-name in cloud service cloud-service-name is not supported because it is in region region-name. Allowed regions: [list of available regions]. | Region is not yet supported for migration. |

-| Migration is not supported for Deployment {0} in HostedService {1} because it uses following features not yet supported for migration: Non-vnet deployment.| Deployment is not located in a virtual network. Refer [this](in-place-migration-technical-details.md#migration-of-deployments-not-in-a-virtual-network) document for more details. |

-| Default VNet destination option not implemented. | ΓÇ£DefaultΓÇ¥ value is not supported for DestinationVirtualNetwork property in the REST request body. |

-| The deployment {0} cannot be migrated because the CSPKG is not available. | Upgrade the deployment and try again. |

-| UnknownExceptionInEndExecute: Contract.Assert failed: rgName is null or empty: Exception received in EndExecute that is not an RdfeException. | Use [Microsoft Q&A](/answers/topics/azure-cloud-services-extended-support.html) or Contact support. |

-| UnknownExceptionInEndExecute: A task was canceled: Exception received in EndExecute that is not an RdfeException. | Use [Microsoft Q&A](/answers/topics/azure-cloud-services-extended-support.html) or Contact support. |

-|XrpPaaSMigrationCscfgCsdefValidationMismatch: Migration could not be completed on deployment deployment-name in hosted service service-name because the deployment's metadata is stale. Please abort the migration and upgrade the deployment before retrying migration. Validation Message: The service name 'service-name'in the service defintion file does not match the name 'service-name-in-config-file' in the service configuration file|match the service names in both .csdef and .cscfg file|

-|NetworkingInternalOperationError when deploying Cloud Service (extended support) resource| The issue may occur if the Service name is same as role name. The reccomended remidiation is to use diferent names for service and roles|

-This guide assumes you've already [created a Computer Vision resource](https://portal.azure.com/#create/Microsoft.CognitiveServicesComputerVision) and obtained a key and endpoint URL.

-> These APIs are only available in the following geographic regions: East US, France Central, Korea Central, North Europe, Southeast Asia, West Europe, West US.

-## Submit data to the service

-When calling the **[Segment](https://centraluseuap.dev.cognitive.microsoft.com/docs/services/unified-vision-apis-public-preview-2023-02-01-preview/operations/63e6b6d9217d201194bbecbd)** API, you specify the image's URL by formatting the request body like this: `{"url":"https://docs.microsoft.com/azure/cognitive-services/computer-vision/images/windows-kitchen.jpg"}`.

-To analyze a local image, you'd put the binary image data in the HTTP request body.

-## Determine how to process the data

-### Select a mode

-|URL parameter |Value |Description |

-||||

-|`mode` | `backgroundRemoval` | Outputs an image of the detected foreground object with a transparent background. |

-|`mode` | `foregroundMatting` | Outputs a grayscale alpha matte image showing the opacity of the detected foreground object. |

-A populated URL for backgroundRemoval would look like this: `https://{endpoint}/computervision/imageanalysis:segment?api-version=2023-02-01-preview&mode=backgroundRemoval`

-This section shows you how to parse the results of the API call.

-The service returns a `200` HTTP response, and the body contains the returned image in the form of a binary stream. The following is an example of the 4-channel PNG image response for the `backgroundRemoval` mode:

-The following is an example of the 1-channel PNG image response for the `foregroundMatting` mode:

-The API will return an image the same size as the original for the `foregroundMatting` mode, but at most 16 megapixels (preserving image aspect ratio) for the `backgroundRemoval` mode.

-See the following list of possible errors and their causes:

- - `Value for mode is invalid.` Ensure you have selected exactly one of the valid options for the `mode` parameter.

- - `This operation is not enabled in this region.` Ensure that your resource is in one of the geographic regions where the API is supported.

- - `The image size is not allowed to be zero or larger than {number} bytes.` Ensure your image is within the specified size limits.

- - `The image dimension is not allowed to be smaller than {min number of pixels} and larger than {max number of pixels}`. Ensure both dimensions of the image are within the specified dimension limits.

- - `Image format is not valid.` Ensure the input data is a valid JPEG, GIF, TIFF, BMP, or PNG image.

- - `InternalServerError.` The processing resulted in an internal error.

- - `ServiceUnavailable.` The service is unavailable.

-This guide assumes you've already <a href="https://portal.azure.com/#create/Microsoft.CognitiveServicesComputerVision" title="created a Computer Vision resource" target="_blank">created a Computer Vision resource </a> and obtained a key and endpoint URL. If you're using a client SDK, you'll also need to authenticate a client object. If you haven't done these steps, follow the [quickstart](../quickstarts-sdk/image-analysis-client-library-40.md) to get started.

-

-## Submit data to the service

-The code in this guide uses remote images referenced by URL. You may want to try different images on your own to see the full capability of the Image Analysis features.

-#### [C#](#tab/csharp)

-In your main class, save a reference to the URL of the image you want to analyze.

-```csharp

-var imageSource = VisionSource.FromUrl(new Uri("https://learn.microsoft.com/azure/cognitive-services/computer-vision/images/windows-kitchen.jpg"));

-```

-> You can also analyze a local image. See the [reference documentation](/dotnet/api/azure.ai.vision.imageanalysis) for alternative **Analyze** methods. Or, see the sample code on [GitHub](https://github.com/Azure-Samples/azure-ai-vision-sdk) for scenarios involving local images.

-#### [Python](#tab/python)

-Save a reference to the URL of the image you want to analyze.

-```python

-image_url = 'https://learn.microsoft.com/azure/cognitive-services/computer-vision/images/windows-kitchen.jpg'

-vision_source = visionsdk.VisionSource(url=image_url)

-```

-> [!TIP]

-> You can also analyze a local image. See the [reference documentation](/python/api/azure-ai-vision) for alternative **Analyze** methods. Or, see the sample code on [GitHub](https://github.com/Azure-Samples/azure-ai-vision-sdk) for scenarios involving local images.

-Save a reference to the URL of the image you want to analyze.

-```cpp

-auto imageSource = VisionSource::FromUrl("https://learn.microsoft.com/azure/cognitive-services/computer-vision/images/windows-kitchen.jpg");

-```

-> [!TIP]

-> You can also analyze a local image. See the [reference documentation]() for alternative **Analyze** methods. Or, see the sample code on [GitHub](https://github.com/Azure-Samples/azure-ai-vision-sdk) for scenarios involving local images.

-#### [REST](#tab/rest)

-When analyzing a remote image, you specify the image's URL by formatting the request body like this: `{"url":"https://learn.microsoft.com/azure/cognitive-services/computer-vision/images/windows-kitchen.jpg"}`.

-To analyze a local image, you'd put the binary image data in the HTTP request body.

-## Determine how to process the data

-### Select visual features

-The Analysis 4.0 API gives you access to all of the service's image analysis features. Choose which operations to do based on your own use case. See the [overview](../overview.md) for a description of each feature. The examples in the sections below add all of the available visual features, but for practical usage you'll likely only need one or two.

-#### [C#](#tab/csharp)

-Define an **ImageAnalysisOptions** object, which specifies visual features you'd like to extract in your analysis.

-```csharp

-var analysisOptions = new ImageAnalysisOptions()

-{

- // Mandatory. You must set one or more features to analyze. Here we use the full set of features.

- // Note that 'Caption' is only supported in Azure GPU regions (East US, France Central, Korea Central,

- // North Europe, Southeast Asia, West Europe, West US and East Asia)

- Features =

- ImageAnalysisFeature.CropSuggestions

- | ImageAnalysisFeature.Caption

- | ImageAnalysisFeature.Objects

- | ImageAnalysisFeature.People

- | ImageAnalysisFeature.Text

- | ImageAnalysisFeature.Tags

-};

-```

-Specify which visual features you'd like to extract in your analysis.

-```python

-# Set the language and one or more visual features as analysis options

-image_analysis_options = visionsdk.ImageAnalysisOptions()

-# Mandatory. You must set one or more features to analyze. Here we use the full set of features.

-# Note that 'Caption' is only supported in Azure GPU regions (East US, France Central, Korea Central,

-# North Europe, Southeast Asia, West Europe, West US)

-image_analysis_options.features = (

- visionsdk.ImageAnalysisFeature.CROP_SUGGESTIONS |

- visionsdk.ImageAnalysisFeature.CAPTION |

- visionsdk.ImageAnalysisFeature.OBJECTS |

- visionsdk.ImageAnalysisFeature.PEOPLE |

- visionsdk.ImageAnalysisFeature.TEXT |

- visionsdk.ImageAnalysisFeature.TAGS

-)

-```

-#### [C++](#tab/cpp)

-Define an **ImageAnalysisOptions** object, which specifies visual features you'd like to extract in your analysis.

-```cpp

-auto analysisOptions = ImageAnalysisOptions::Create();

-analysisOptions->SetFeatures(

- {

- ImageAnalysisFeature::CropSuggestions,

- ImageAnalysisFeature::Caption,

- ImageAnalysisFeature::Objects,

- ImageAnalysisFeature::People,

- ImageAnalysisFeature::Text,

- ImageAnalysisFeature::Tags

- });

-```

-#### [REST](#tab/rest)

-You can specify which features you want to use by setting the URL query parameters of the [Analysis 4.0 API](https://aka.ms/vision-4-0-ref). A parameter can have multiple values, separated by commas. Each feature you specify will require more computation time, so only specify what you need.

-|URL parameter | Value | Description|

-|||--|

-|`features`|`Read` | reads the visible text in the image and outputs it as structured JSON data.|

-|`features`|`Caption` | describes the image content with a complete sentence in supported languages.|

-|`features`|`DenseCaption` | generates detailed captions for up to 10 prominent image regions. |

-|`features`|`SmartCrops` | finds the rectangle coordinates that would crop the image to a desired aspect ratio while preserving the area of interest.|

-|`features`|`Objects` | detects various objects within an image, including the approximate location. The Objects argument is only available in English.|

-|`features`|`Tags` | tags the image with a detailed list of words related to the image content.|

-A populated URL might look like this:

-`https://{endpoint}/computervision/imageanalysis:analyze?api-version=2023-02-01-preview&features=tags,read,caption,denseCaption,smartCrops,objects,people`

-### Use a custom model

-You can also do image analysis with a custom trained model. To create and train a model, see [Create a custom Image Analysis model](./model-customization.md). Once your model is trained, all you need is the model's name value.

-#### [C#](#tab/csharp)

-To use a custom model, create the ImageAnalysisOptions with no features, and set the name of your model.

-```csharp

-var analysisOptions = new ImageAnalysisOptions()

-{

- ModelName = "MyCustomModelName"

-};

-```

-#### [Python](#tab/python)

-To use a custom model, create an **ImageAnalysisOptions** object with no features set, and set the name of your model.

-```python

-analysis_options = sdk.ImageAnalysisOptions()

-analysis_options.model_name = "MyCustomModelName"

-```

-To use a custom model, create an **ImageAnalysisOptions** object with no features set, and set the name of your model.

-```cpp

-auto analysisOptions = ImageAnalysisOptions::Create();

-analysisOptions->SetModelName("MyCustomModelName");

-```

-#### [REST](#tab/rest)

-To use a custom model, do not use the features query parameter. Set the model-name parameter to the name of your model.

-`https://{endpoint}/computervision/imageanalysis:analyze?api-version=2023-02-01-preview&model-name=MyCustomModelName`

-You can also specify the language of the returned data. This is optional, and the default language is English. See [Language support](https://aka.ms/cv-languages) for a list of supported language codes and which visual features are supported for each language.

-Use the *language* property of your **ImageAnalysisOptions** object to specify a language.

-```csharp

-var analysisOptions = new ImageAnalysisOptions()

-{

- // Optional. Default is "en" for English. See https://aka.ms/cv-languages for a list of supported

- // language codes and which visual features are supported for each language.

- Language = "en",

-};

-```

-Use the *language* property of your **ImageAnalysisOptions** object to specify a language.

-```python

-# Optional. Default is "en" for English. See https://aka.ms/cv-languages for a list of supported

-# language codes and which visual features are supported for each language.

-image_analysis_options.language = 'en'

-```

-Use the *language* property of your **ImageAnalysisOptions** object to specify a language.

-```cpp

-// Optional. Default is "en" for English. See https://aka.ms/cv-languages for a list of supported

-// language codes and which visual features are supported for each language.

-analysisOptions->SetLanguage("en");

-```

-#### [REST](#tab/rest)

-`https://{endpoint}/computervision/imageanalysis:analyze?api-version=2023-02-01-preview&features=tags,read,caption,denseCaption,smartCrops,objects,people&language=en`

-## Get results from the service

-This section shows you how to parse the results of the API call. It includes the API call itself.

-### With visual features

-The following code calls the Image Analysis API and prints the results for all standard visual features.

-```csharp

-using var analyzer = new ImageAnalyzer(serviceOptions, imageSource, analysisOptions);

-var result = analyzer.Analyze();

-if (result.Reason == ImageAnalysisResultReason.Analyzed)

-{

- Console.WriteLine($" Image height = {result.ImageHeight}");

- Console.WriteLine($" Image width = {result.ImageWidth}");

- Console.WriteLine($" Model version = {result.ModelVersion}");

- if (result.Caption != null)

- {

- Console.WriteLine(" Caption:");

- Console.WriteLine($" \"{result.Caption.Content}\", Confidence {result.Caption.Confidence:0.0000}");

- }

- if (result.Objects != null)

- {

- Console.WriteLine(" Objects:");

- foreach (var detectedObject in result.Objects)

- {

- Console.WriteLine($" \"{detectedObject.Name}\", Bounding box {detectedObject.BoundingBox}, Confidence {detectedObject.Confidence:0.0000}");

- }

- }

- if (result.Tags != null)

- {

- Console.WriteLine($" Tags:");

- foreach (var tag in result.Tags)

- {

- Console.WriteLine($" \"{tag.Name}\", Confidence {tag.Confidence:0.0000}");

- }

- }

- if (result.People != null)

- {

- Console.WriteLine($" People:");

- foreach (var person in result.People)

- {

- Console.WriteLine($" Bounding box {person.BoundingBox}, Confidence {person.Confidence:0.0000}");

- }

- }

- if (result.CropSuggestions != null)

- {

- Console.WriteLine($" Crop Suggestions:");

- foreach (var cropSuggestion in result.CropSuggestions)

- {

- Console.WriteLine($" Aspect ratio {cropSuggestion.AspectRatio}: "

- + $"Crop suggestion {cropSuggestion.BoundingBox}");

- };

- }

- if (result.Text != null)

- {

- Console.WriteLine($" Text:");

- foreach (var line in result.Text.Lines)

- {

- string pointsToString = "{" + string.Join(',', line.BoundingPolygon.Select(pointsToString => pointsToString.ToString())) + "}";

- Console.WriteLine($" Line: '{line.Content}', Bounding polygon {pointsToString}");

- foreach (var word in line.Words)

- {

- pointsToString = "{" + string.Join(',', word.BoundingPolygon.Select(pointsToString => pointsToString.ToString())) + "}";

- Console.WriteLine($" Word: '{word.Content}', Bounding polygon {pointsToString}, Confidence {word.Confidence:0.0000}");

- }

- }

- }

- var resultDetails = ImageAnalysisResultDetails.FromResult(result);

- Console.WriteLine($" Result details:");

- Console.WriteLine($" Image ID = {resultDetails.ImageId}");

- Console.WriteLine($" Result ID = {resultDetails.ResultId}");

- Console.WriteLine($" Connection URL = {resultDetails.ConnectionUrl}");

- Console.WriteLine($" JSON result = {resultDetails.JsonResult}");

-}

-else if (result.Reason == ImageAnalysisResultReason.Error)

-{

- var errorDetails = ImageAnalysisErrorDetails.FromResult(result);

- Console.WriteLine(" Analysis failed.");

- Console.WriteLine($" Error reason : {errorDetails.Reason}");

- Console.WriteLine($" Error code : {errorDetails.ErrorCode}");

- Console.WriteLine($" Error message: {errorDetails.Message}");

-}

-```

-### With custom model

-The following code calls the Image Analysis API and prints the results for custom model analysis.

-```csharp

-using var analyzer = new ImageAnalyzer(serviceOptions, imageSource, analysisOptions);

-var result = analyzer.Analyze();

-if (result.Reason == ImageAnalysisResultReason.Analyzed)

-{

- if (result.CustomObjects != null)

- {

- Console.WriteLine(" Custom Objects:");

- foreach (var detectedObject in result.CustomObjects)

- {

- Console.WriteLine($" \"{detectedObject.Name}\", Bounding box {detectedObject.BoundingBox}, Confidence {detectedObject.Confidence:0.0000}");

- }

- }

- if (result.CustomTags != null)

- {

- Console.WriteLine($" Custom Tags:");

- foreach (var tag in result.CustomTags)

- {

- Console.WriteLine($" \"{tag.Name}\", Confidence {tag.Confidence:0.0000}");

- }

- }

-}

-else if (result.Reason == ImageAnalysisResultReason.Error)

-{

- var errorDetails = ImageAnalysisErrorDetails.FromResult(result);

- Console.WriteLine(" Analysis failed.");

- Console.WriteLine($" Error reason : {errorDetails.Reason}");

- Console.WriteLine($" Error code : {errorDetails.ErrorCode}");

- Console.WriteLine($" Error message: {errorDetails.Message}");

-}

-```

-### With visual features

-The following code calls the Image Analysis API and prints the results for all standard visual features.

-```python

-image_analyzer = sdk.ImageAnalyzer(service_options, vision_source, analysis_options)

-result = image_analyzer.analyze()

-if result.reason == sdk.ImageAnalysisResultReason.ANALYZED:

- print(" Image height: {}".format(result.image_height))

- print(" Image width: {}".format(result.image_width))

- print(" Model version: {}".format(result.model_version))

- if result.caption is not None:

- print(" Caption:")

- print(" '{}', Confidence {:.4f}".format(result.caption.content, result.caption.confidence))

- if result.objects is not None:

- print(" Objects:")

- for object in result.objects:

- print(" '{}', {} Confidence: {:.4f}".format(object.name, object.bounding_box, object.confidence))

- if result.tags is not None:

- print(" Tags:")

- for tag in result.tags:

- print(" '{}', Confidence {:.4f}".format(tag.name, tag.confidence))

- if result.people is not None:

- print(" People:")

- for person in result.people:

- print(" {}, Confidence {:.4f}".format(person.bounding_box, person.confidence))

- if result.crop_suggestions is not None:

- print(" Crop Suggestions:")

- for crop_suggestion in result.crop_suggestions:

- print(" Aspect ratio {}: Crop suggestion {}"

- .format(crop_suggestion.aspect_ratio, crop_suggestion.bounding_box))

- if result.text is not None:

- print(" Text:")

- for line in result.text.lines:

- points_string = "{" + ", ".join([str(int(point)) for point in line.bounding_polygon]) + "}"

- print(" Line: '{}', Bounding polygon {}".format(line.content, points_string))

- for word in line.words:

- points_string = "{" + ", ".join([str(int(point)) for point in word.bounding_polygon]) + "}"

- print(" Word: '{}', Bounding polygon {}, Confidence {:.4f}"

- .format(word.content, points_string, word.confidence))

- result_details = sdk.ImageAnalysisResultDetails.from_result(result)

- print(" Result details:")

- print(" Image ID: {}".format(result_details.image_id))

- print(" Result ID: {}".format(result_details.result_id))

- print(" Connection URL: {}".format(result_details.connection_url))

- print(" JSON result: {}".format(result_details.json_result))

-elif result.reason == sdk.ImageAnalysisResultReason.ERROR:

- error_details = sdk.ImageAnalysisErrorDetails.from_result(result)

- print(" Analysis failed.")

- print(" Error reason: {}".format(error_details.reason))

- print(" Error code: {}".format(error_details.error_code))

- print(" Error message: {}".format(error_details.message))

-```

-### With custom model

-The following code calls the Image Analysis API and prints the results for custom model analysis.

-```python

-image_analyzer = sdk.ImageAnalyzer(service_options, vision_source, analysis_options)

-result = image_analyzer.analyze()

-if result.reason == sdk.ImageAnalysisResultReason.ANALYZED:

- if result.custom_objects is not None:

- print(" Custom Objects:")

- for object in result.custom_objects:

- print(" '{}', {} Confidence: {:.4f}".format(object.name, object.bounding_box, object.confidence))

- if result.custom_tags is not None:

- print(" Custom Tags:")

- for tag in result.custom_tags:

- print(" '{}', Confidence {:.4f}".format(tag.name, tag.confidence))

-elif result.reason == sdk.ImageAnalysisResultReason.ERROR:

- error_details = sdk.ImageAnalysisErrorDetails.from_result(result)

- print(" Analysis failed.")

- print(" Error reason: {}".format(error_details.reason))

- print(" Error code: {}".format(error_details.error_code))

- print(" Error message: {}".format(error_details.message))

-```

-#### [C++](#tab/cpp)

-### With visual features

-The following code calls the Image Analysis API and prints the results for all standard visual features.

-```cpp

-auto analyzer = ImageAnalyzer::Create(serviceOptions, imageSource, analysisOptions);

-auto result = analyzer->Analyze();

-if (result->GetReason() == ImageAnalysisResultReason::Analyzed)

-{

- std::cout << " Image height = " << result->GetImageHeight().Value() << std::endl;

- std::cout << " Image width = " << result->GetImageWidth().Value() << std::endl;

- std::cout << " Model version = " << result->GetModelVersion().Value() << std::endl;

- const auto caption = result->GetCaption();

- if (caption.HasValue())

- {

- std::cout << " Caption:" << std::endl;

- std::cout << " \"" << caption.Value().Content << "\", Confidence " << caption.Value().Confidence << std::endl;

- }

- const auto objects = result->GetObjects();

- if (objects.HasValue())

- {

- std::cout << " Objects:" << std::endl;

- for (const auto object : objects.Value())

- {

- std::cout << " \"" << object.Name << "\", ";

- std::cout << "Bounding box " << object.BoundingBox.ToString();

- std::cout << ", Confidence " << object.Confidence << std::endl;

- }

- }

- const auto tags = result->GetTags();

- if (tags.HasValue())

- {

- std::cout << " Tags:" << std::endl;

- for (const auto tag : tags.Value())

- {

- std::cout << " \"" << tag.Name << "\"";

- std::cout << ", Confidence " << tag.Confidence << std::endl;

- }

- }

- const auto people = result->GetPeople();

- if (people.HasValue())

- {

- std::cout << " People:" << std::endl;

- for (const auto person : people.Value())

- {

- std::cout << " Bounding box " << person.BoundingBox.ToString();

- std::cout << ", Confidence " << person.Confidence << std::endl;

- }

- }

- const auto cropSuggestions = result->GetCropSuggestions();

- if (cropSuggestions.HasValue())

- {

- std::cout << " Crop Suggestions:" << std::endl;

- for (const auto cropSuggestion : cropSuggestions.Value())

- {

- std::cout << " Aspect ratio " << cropSuggestion.AspectRatio;

- std::cout << ": Crop suggestion " << cropSuggestion.BoundingBox.ToString() << std::endl;

- }

- }

- const auto detectedText = result->GetText();

- if (detectedText.HasValue())

- {

- std::cout << " Text:\n";

- for (const auto line : detectedText.Value().Lines)

- {

- std::cout << " Line: \"" << line.Content << "\"";

- std::cout << ", Bounding polygon " << PolygonToString(line.BoundingPolygon) << std::endl;

- for (const auto word : line.Words)

- {

- std::cout << " Word: \"" << word.Content << "\"";

- std::cout << ", Bounding polygon " << PolygonToString(word.BoundingPolygon);

- std::cout << ", Confidence " << word.Confidence << std::endl;

- }

- }

- }

- auto resultDetails = ImageAnalysisResultDetails::FromResult(result);

- std::cout << " Result details:\n";;

- std::cout << " Image ID = " << resultDetails->GetImageId() << std::endl;

- std::cout << " Result ID = " << resultDetails->GetResultId() << std::endl;

- std::cout << " Connection URL = " << resultDetails->GetConnectionUrl() << std::endl;

- std::cout << " JSON result = " << resultDetails->GetJsonResult() << std::endl;

-}

-else if (result->GetReason() == ImageAnalysisResultReason::Error)

-{

- auto errorDetails = ImageAnalysisErrorDetails::FromResult(result);

- std::cout << " Analysis failed." << std::endl;

- std::cout << " Error reason = " << (int)errorDetails->GetReason() << std::endl;

- std::cout << " Error code = " << errorDetails->GetErrorCode() << std::endl;

- std::cout << " Error message = " << errorDetails->GetMessage() << std::endl;

-}

-```

-Use the following helper method to display rectangle coordinates:

-```cpp

-std::string PolygonToString(std::vector<int32_t> boundingPolygon)

-{

- std::string out = "{";

- for (int i = 0; i < boundingPolygon.size(); i += 2)

- {

- out += ((i == 0) ? "{" : ",{") +

- std::to_string(boundingPolygon[i]) + "," +

- std::to_string(boundingPolygon[i + 1]) + "}";

- }

- out += "}";

- return out;

-}

-```

-### With custom model

-The following code calls the Image Analysis API and prints the results for custom model analysis.

-```cpp

-auto analyzer = ImageAnalyzer::Create(serviceOptions, imageSource, analysisOptions);

-auto result = analyzer->Analyze();

-if (result->GetReason() == ImageAnalysisResultReason::Analyzed)

-{

- const auto objects = result->GetCustomObjects();

- if (objects.HasValue())

- {

- std::cout << " Custom objects:" << std::endl;

- for (const auto object : objects.Value())

- {

- std::cout << " \"" << object.Name << "\", ";

- std::cout << "Bounding box " << object.BoundingBox.ToString();

- std::cout << ", Confidence " << object.Confidence << std::endl;

- }

- }

- const auto tags = result->GetCustomTags();

- if (tags.HasValue())

- {

- std::cout << " Custom tags:" << std::endl;

- for (const auto tag : tags.Value())

- {

- std::cout << " \"" << tag.Name << "\"";

- std::cout << ", Confidence " << tag.Confidence << std::endl;

- }

- }

-}

-else if (result->GetReason() == ImageAnalysisResultReason::Error)

-{

- auto errorDetails = ImageAnalysisErrorDetails::FromResult(result);

- std::cout << " Analysis failed." << std::endl;

- std::cout << " Error reason = " << (int)errorDetails->GetReason() << std::endl;

- std::cout << " Error code = " << errorDetails->GetErrorCode() << std::endl;

- std::cout << " Error message = " << errorDetails->GetMessage() << std::endl;

-}

-```

-#### [REST](#tab/rest)

-### Error codes

-See the following list of possible errors and their causes:

-* 400

- * `InvalidImageUrl` - Image URL is badly formatted or not accessible.

- * `InvalidImageFormat` - Input data is not a valid image.

- * `InvalidImageSize` - Input image is too large.

- * `NotSupportedVisualFeature` - Specified feature type isn't valid.

- * `NotSupportedImage` - Unsupported image, for example child pornography.

- * `InvalidDetails` - Unsupported `detail` parameter value.

- * `NotSupportedLanguage` - The requested operation isn't supported in the language specified.

- * `BadArgument` - More details are provided in the error message.

-* 415 - Unsupported media type error. The Content-Type isn't in the allowed types:

- * For an image URL, Content-Type should be `application/json`

- * For a binary image data, Content-Type should be `application/octet-stream` or `multipart/form-data`

-* 500

- * `FailedToProcess`

- * `Timeout` - Image processing timed out.

- * `InternalServerError`

-> [!TIP]

-> While working with Computer Vision, you might encounter transient failures caused by [rate limits](https://azure.microsoft.com/pricing/details/cognitive-services/computer-vision/) enforced by the service, or other transient problems like network outages. For information about handling these types of failures, see [Retry pattern](/azure/architecture/patterns/retry) in the Cloud Design Patterns guide, and the related [Circuit Breaker pattern](/azure/architecture/patterns/circuit-breaker).

-* Explore the [code samples on GitHub](https://github.com/Azure-Samples/azure-ai-vision-sdk/blob/main/samples/).

-* See the [API reference](https://aka.ms/vision-4-0-ref) to learn more about the API functionality.

-`https://{endpoint}/computervision/imageanalysis:analyze?api-version=2023-02-01-preview&features=Tags`

-`https://{endpoint}/computervision/imageanalysis:analyze?api-version=2023-02-01-preview&features=Tags&language=en`

-> Some of the features in Image Analysis can be called directly as well as through the Analyze API call. For example, you can do a scoped analysis of only image tags by making a request to `https://{endpoint}/vision/v3.2/tag` (or to the corresponding method in the SDK). See the [reference documentation](https://westus.dev.cognitive.microsoft.com/docs/services/computer-vision-v3-2/operations/56f91f2e778daf14a499f21b) for other features that can be called separately.

-This guide shows you how to create and train a custom image classification model. The few differences between this and object detection models are noted.

-* [Call the Analyze Image API](./call-analyze-image-40.md#use-a-custom-model)

-Get started with the Image Analysis 4.0 REST API or client libraries to set up a basic image analysis application. The Image Analysis service provides you with AI algorithms for processing images and returning information on their visual features. Follow these steps to install a package to your application and try out the sample code.

-* **Use all data forms**: If your LUIS app will take data in multiple forms, make sure to include those forms in your training utterances. For example, if your client application takes both speech and typed text input, you need to have speech-to-text generated utterances as well as typed utterances. You will see different variations in how people speak from how they type as well as different errors in speech recognition and typos. All of this variation should be represented in your training data.

-To create a transcription, use the [Transcriptions_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Transcriptions_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md#transcriptions). Construct the request body according to the following instructions:

-|`diarization`|Indicates that diarization analysis should be carried out on the input, which is expected to be a mono channel that contains multiple voices. Specify the minimum and maximum number of people who might be speaking. You must also set the `diarizationEnabled` property to `true`. The [transcription file](batch-transcription-get.md#transcription-result-file) will contain a `speaker` entry for each transcribed phrase.<br/><br/>You need to use this property when you expect three or more speakers. For two speakers setting `diarizationEnabled` property to `true` is enough. See an example of the property usage in [Transcriptions_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Transcriptions_Create) operation description.<br/><br/>Diarization is the process of separating speakers in audio data. The batch pipeline can recognize and separate multiple speakers on mono channel recordings. The feature isn't available with stereo recordings.<br/><br/>When this property is selected, source audio length can't exceed 240 minutes per file.<br/><br/>**Note**: This property is only available with Speech-to-text REST API version 3.1.|

-|`diarizationEnabled`|Specifies that diarization analysis should be carried out on the input, which is expected to be a mono channel that contains two voices. The default value is `false`.<br/><br/>For three or more voices you also need to use property `diarization` (only with Speech-to-text REST API version 3.1).<br/><br/>When this property is selected, source audio length can't exceed 240 minutes per file.|

-To get the status of the transcription job, call the [Transcriptions_Get](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Transcriptions_Get) operation of the [Speech-to-text REST API](rest-speech-to-text.md).

-|`displayPhraseElements`|A list of results with display text for each word of the phrase. The `displayFormWordLevelTimestampsEnabled` request property must be set to `true`, otherwise this property is not present.<br/><br/>**Note**: This property is only available with speech-to-text REST API version 3.1.|

-|`locale`|The locale identified from the input the audio. The `languageIdentification` request property must be set, otherwise this property is not present.<br/><br/>**Note**: This property is only available with speech-to-text REST API version 3.1.|

-Batch transcription is used to transcribe a large amount of audio data in storage. Both the [Speech-to-text REST API](rest-speech-to-text.md#transcriptions) and [Speech CLI](spx-basics.md) support batch transcription.

-> For a low or no-code solution, you can use the [Batch Speech-to-text Connector](/connectors/cognitiveservicesspe/) in Power Platform applications such as Power Automate, Power Apps, and Logic Apps. See the [Power automate batch transcription](power-automate-batch-transcription.md) guide to get started.

-| [Custom Speech](./custom-speech-overview.md) | A speech-to-text feature used evaluate and improve the speech recognition accuracy of use-case specific entities (such as alpha-numeric customer, case, and contract IDs, license plates, and names). You can also train a custom model with your own product names and industry terminology. |

-| [Custom Neural Voice](./custom-neural-voice.md) | A text-to-speech feature that lets you create a one-of-a-kind, customized, synthetic voice for your applications. |

-description: A common scenario for speech-to-text is transcribing large volumes of telephony data that come from various systems, such as interactive voice response (IVR) in real-time. This requires an integration with the Telephony System used.

-> For guidance on reducing Text to Speech latency check out the **[How to lower speech synthesis latency](./how-to-lower-speech-synthesis-latency.md?pivots=programming-language-csharp)** guide.

-> In addition, consider implementing a Text to Speech cache to store all synthesized audio and playback from the cache in case a string has previously been synthesized.

-Conversation transcription is a [speech-to-text](speech-to-text.md) solution that provides real-time or asynchronous transcription of any conversation. This feature, which is currently in preview, combines speech recognition, speaker identification, and sentence attribution to determine who said what, and when, in a conversation.

-Currently, conversation transcription supports [all speech-to-text languages](language-support.md?tabs=stt) in the following regions:ΓÇ»`centralus`, `eastasia`, `eastus`, `westeurope`.

-Applications such as [Voice assistants](voice-assistants.md) listen to users and take an action in response, often speaking back. They use [speech-to-text](speech-to-text.md) to transcribe the user's speech, then take action on the natural language understanding of the text. This action frequently includes spoken output from the assistant generated with [text-to-speech](text-to-speech.md). Devices connect to assistants with the Speech SDK's `DialogServiceConnector` object.

-1. Select the voice model name to review the model details and listen to the sample text-to-speech results.

-* [Try the text-to-speech quickstart](get-started-text-to-speech.md)

-description: Custom Neural Voice is a text-to-speech feature that allows you to create a one-of-a-kind, customized, synthetic voice for your applications. You provide your own audio data as a sample.

-Custom Neural Voice (CNV) is a text-to-speech feature that lets you create a one-of-a-kind, customized, synthetic voice for your applications. With Custom Neural Voice, you can build a highly natural-sounding voice for your brand or characters by providing human speech samples as training data.

-Out of the box, [text-to-speech](text-to-speech.md) can be used with prebuilt neural voices for each [supported language](language-support.md?tabs=tts). The prebuilt neural voices work very well in most text-to-speech scenarios if a unique voice isn't required.

-Custom Neural Voice is based on the neural text-to-speech technology and the multilingual, multi-speaker, universal model. You can create synthetic voices that are rich in speaking styles, or adaptable cross languages. The realistic and natural sounding voice of Custom Neural Voice can represent brands, personify machines, and allow users to interact with applications conversationally. See the [supported languages](language-support.md?tabs=tts) for Custom Neural Voice.

-The style and the characteristics of the trained voice model depend on the style and the quality of the recordings from the voice talent used for training. However, you can make several adjustments by using [SSML (Speech Synthesis Markup Language)](./speech-synthesis-markup.md?tabs=csharp) when you make the API calls to your voice model to generate synthetic speech. SSML is the markup language used to communicate with the text-to-speech service to convert text into audio. The adjustments you can make include change of pitch, rate, intonation, and pronunciation correction. If the voice model is built with multiple styles, you can also use SSML to switch the styles.

-Neural text-to-speech voice models are trained by using deep neural networks based on

-* [Code of Conduct for Text-to-Speech integrations](/legal/cognitive-services/speech-service/tts-code-of-conduct?context=/azure/cognitive-services/speech-service/context/context)

-description: Custom Speech is a set of online tools that allows you to evaluate and improve the Microsoft speech-to-text accuracy for your applications, tools, and products.

-With Custom Speech, you can evaluate and improve the accuracy of speech recognition for your applications and products. A custom speech model can be used for [real-time speech-to-text](speech-to-text.md), [speech translation](speech-translation.md), and [batch transcription](batch-transcription.md).

-1. [Upload test data](./how-to-custom-speech-upload-data.md). Upload test data to evaluate the Microsoft speech-to-text offering for your applications, tools, and products.

-1. [Test model quantitatively](how-to-custom-speech-evaluate-data.md). Evaluate and improve the accuracy of the speech-to-text model. The Speech service provides a quantitative word error rate (WER), which you can use to determine if additional training is required.

-description: Use phonemes to customize pronunciation of words in Speech-to-Text.

-[Voice assistants](voice-assistants.md) listen to users and take an action in response, often speaking back. They use [speech-to-text](speech-to-text.md) to transcribe the user's speech, then take action on the natural language understanding of the text. This action frequently includes spoken output from the assistant generated with [text-to-speech](text-to-speech.md).

-Voice assistants built using Speech service can use the full range of customization options available for [speech-to-text](speech-to-text.md), [text-to-speech](text-to-speech.md), and [custom keyword selection](./custom-keyword-basics.md).

-Speech-to-text offers an array of formatting features to ensure that the transcribed text is clear and legible. Below is an overview of these features and how each one is used to improve the overall clarity of the final text output.

-Inverse Text Normalization (ITN) is a process that converts spoken words into their written form. For example, the spoken word "four" is converted to the written form "4". This process is performed by the speech-to-text service and isn't configurable. Some of the supported text formats include dates, times, decimals, currencies, addresses, emails, and phone numbers. You can speak naturally, and the service formats text as expected. The following table shows the ITN rules that are applied to the text output.

-Speech-to-text models recognize words that should be capitalized to improve readability, accuracy, and grammar. For example, the Speech service will automatically capitalize proper nouns and words at the beginning of a sentence. Some examples are shown in this table.

-When speaking, it's common for someone to stutter, duplicate words, and say filler words like "uhm" or "uh". Speech-to-text can recognize such disfluencies and remove them from the display text. Disfluency removal is great for transcribing live unscripted speeches to read them back later. Some examples are shown in this table.

-Speech-to-text automatically punctuates your text to improve clarity. Punctuation is helpful for reading back call or conversation transcriptions. Some examples are shown in this table.

-When you're using speech-to-text with continuous recognition, you can configure the Speech service to recognize explicit punctuation marks. Then you can speak punctuation aloud in order to make your text more legible. This is especially useful in a situation where you want to use complex punctuation without having to merge it later. Some examples are shown in this table.

-Use the Speech SDK to enable dictation mode when you're using speech-to-text with continuous recognition. This mode will cause the speech configuration instance to interpret word descriptions of sentence structures such as punctuation.

-* [Speech-to-text quickstart](get-started-speech-to-text.md)

-Embedded Speech is designed for on-device [speech-to-text](speech-to-text.md) and [text-to-speech](text-to-speech.md) scenarios where cloud connectivity is intermittent or unavailable. For example, you can use embedded speech in industrial equipment, a voice enabled air conditioning unit, or a car that might travel out of range. You can also develop hybrid cloud and offline solutions. For scenarios where your devices must be in a secure environment like a bank or government entity, you should first consider [disconnected containers](../containers/disconnected-containers.md).

-For embedded speech, you'll need to download the speech recognition models for [speech-to-text](speech-to-text.md) and voices for [text-to-speech](text-to-speech.md). Instructions will be provided upon successful completion of the [limited access review](https://aka.ms/csgate-embedded-speech) process.

-The following [speech-to-text](speech-to-text.md) models are available: de-DE, en-AU, en-CA, en-GB, en-IE, en-IN, en-NZ, en-US, es-ES, es-MX, fr-CA, fr-FR, hi-IN, it-IT, ja-JP, ko-KR, nl-NL, pt-BR, ru-RU, sv-SE, tr-TR, zh-CN, zh-HK, and zh-TW.

-The following [text-to-speech](text-to-speech.md) locales and voices are available:

-| Locale (BCP-47) | Language | Text-to-speech voices |

-Use the `EmbeddedSpeechConfig` object to set the location of the models or voices. If your application is used for both speech-to-text and text-to-speech, you can use the same `EmbeddedSpeechConfig` object to set the location of the models and voices.

-// For speech-to-text

-// For text-to-speech

-> The `GetEnvironmentVariable` function is defined in the [speech-to-text quickstart](get-started-speech-to-text.md) and [text-to-speech quickstart](get-started-text-to-speech.md).

-// For speech-to-text

-// For text-to-speech

-// For speech-to-text

-// For text-to-speech

-With hybrid speech configuration for [speech-to-text](speech-to-text.md) (recognition models), embedded speech is used when connection to the cloud service fails after repeated attempts. Recognition may continue using the cloud service again if the connection is later resumed.

-With hybrid speech configuration for [text-to-speech](text-to-speech.md) (voices), embedded and cloud synthesis are run in parallel and the result is selected based on which one gives a faster response. The best result is evaluated on each synthesis request.

-For cloud speech, you use the `SpeechConfig` object, as shown in the [speech-to-text quickstart](get-started-speech-to-text.md) and [text-to-speech quickstart](get-started-text-to-speech.md). To run the quickstarts for embedded speech, you can replace `SpeechConfig` with `EmbeddedSpeechConfig` or `HybridSpeechConfig`. Most of the other speech recognition and synthesis code are the same, whether using cloud, embedded, or hybrid configuration.

-You can use the [Speech SDK](speech-sdk.md) or [Speech CLI](spx-overview.md) for real-time low latency speech-to-text, text-to-speech, language identification, and speech translation. You can also use the [Batch transcription API](batch-transcription.md) to transcribe pre-recorded speech to text. To synthesize a large volume of text input (long and short) to speech, use the [Batch synthesis API](batch-synthesis.md).

-## Text-to-speech

-Help bring everyone into the conversation by converting text messages to audio using [Text-to-Speech](text-to-speech.md) for scenarios, such as game dialogue prototyping, greater accessibility, or non-playable character (NPC) voices. Text-to-Speech includes [prebuilt neural voice](language-support.md?tabs=tts#prebuilt-neural-voices) and [custom neural voice](language-support.md?tabs=tts#custom-neural-voice) features. Prebuilt neural voice can provide highly natural out-of-box voices with leading voice variety in terms of a large portfolio of languages and voices. Custom neural voice is an easy-to-use self-service for creating a highly natural custom voice.

-For an example, see the [Text-to-speech quickstart](get-started-text-to-speech.md).

-## Speech-to-text

-You can use [speech-to-text](speech-to-text.md) to display text from the spoken audio in your game. For an example, see the [Speech-to-text quickstart](get-started-speech-to-text.md).

-* [Text-to-speech quickstart](get-started-text-to-speech.md)

-* [Speech-to-text quickstart](get-started-speech-to-text.md)

-1. Choose the voice and the language for your script content. Audio Content Creation includes all of the [prebuilt text-to-speech voices](language-support.md?tabs=tts). You can use prebuilt neural voices or a custom neural voice.

- For more information about fine-tuning speech output, view the [How to convert Text to Speech using Microsoft Azure AI voices](https://youtu.be/ygApYuOOG6w) video.

- NotifyUser($"Final speech-to-text result: '{recognitionEventArgs.Result.Text}'");

-To create a project, use the [Projects_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Projects_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To create an endpoint and deploy a model, use the [Endpoints_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Endpoints_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To redeploy the custom endpoint with a new model, use the [Endpoints_Update](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Endpoints_Update) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To get logs for an endpoint, start by using the [Endpoints_Get](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Endpoints_Get) operation of the [Speech-to-text REST API](rest-speech-to-text.md).

-description: In this article, you learn how to quantitatively measure and improve the quality of our speech-to-text model or your custom model.

-In this article, you learn how to quantitatively measure and improve the accuracy of the Microsoft speech-to-text model or your own custom models. [Audio + human-labeled transcript](how-to-custom-speech-test-and-train.md#audio--human-labeled-transcript-data-for-training-or-testing) data is required to test accuracy. You should provide from 30 minutes to 5 hours of representative audio.

-You can test the accuracy of your custom model by creating a test. A test requires a collection of audio files and their corresponding transcriptions. You can compare a custom model's accuracy with a Microsoft speech-to-text base model or another custom model. After you [get](#get-test-results) the test results, [evaluate](#evaluate-word-error-rate) the word error rate (WER) compared to speech recognition results.

-To create a test, use the [Evaluations_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Evaluations_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To get test results, start by using the [Evaluations_Get](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Evaluations_Get) operation of the [Speech-to-text REST API](rest-speech-to-text.md).

-To create a test, use the [Evaluations_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Evaluations_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To get test results, start by using the [Evaluations_Get](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Evaluations_Get) operation of the [Speech-to-text REST API](rest-speech-to-text.md).

-To get the training and transcription expiration dates for a base model, use the [Models_GetBaseModel](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Models_GetBaseModel) operation of the [Speech-to-text REST API](rest-speech-to-text.md). You can make a [Models_ListBaseModels](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Models_ListBaseModels) request to get available base models for all locales.

-To get the transcription expiration date for your custom model, use the [Models_GetCustomModel](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Models_GetCustomModel) operation of the [Speech-to-text REST API](rest-speech-to-text.md).

-Audio data is optimal for testing the accuracy of Microsoft's baseline speech-to-text model or a custom model. Keep in mind that audio data is used to inspect the accuracy of speech with regard to a specific model's performance. If you want to quantify the accuracy of a model, use [audio + human-labeled transcripts](#audio--human-labeled-transcript-data-for-training-or-testing).

-description: Learn how to train Custom Speech models. Training a speech-to-text model can improve recognition accuracy for the Microsoft base model or a custom model.

-To create a model with datasets for training, use the [Models_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Models_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-Copying a model directly to a project in another region is not supported with the Speech CLI. You can copy a model to a project in another region using the [Speech Studio](https://aka.ms/speechstudio/customspeech) or [Speech-to-text REST API](rest-speech-to-text.md).

-To copy a model to another Speech resource, use the [Models_CopyTo](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Models_CopyTo) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To connect a new model to a project of the Speech resource where the model was copied, use the [Models_Update](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Models_Update) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-To create a dataset and connect it to an existing project, use the [Datasets_Create](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Datasets_Create) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-Azure Text-to-Speech engines are updated from time to time to capture the latest language model that defines the pronunciation of the language. After you've trained your voice, you can apply your voice to the new language model by updating to the latest engine version.

-When you're ready to create a custom Text-to-Speech voice for your application, the first step is to gather audio recordings and associated scripts to start training the voice model. For details on recording voice samples, see [the tutorial](record-custom-voice-samples.md). The Speech service uses this data to create a unique voice tuned to match the voice in the recordings. After you've trained the voice, you can start synthesizing speech in your applications.

-When you're ready to create a custom Text-to-Speech voice for your application, the first step is to gather audio recordings and associated scripts to start training the voice model. The Speech service uses this data to create a unique voice tuned to match the voice in the recordings. After you've trained the voice, you can start synthesizing speech in your applications.

-This table lists data types and how each is used to create a custom Text-to-Speech voice model.

-Each audio file should contain a single utterance (a single sentence or a single turn of a dialog system), less than 15 seconds long. All files must be in the same spoken language. Multi-language custom Text-to-Speech voices aren't supported, except for the Chinese-English bi-lingual. Each audio file must have a unique filename with the filename extension .wav.

-In some cases, you may not have segmented audio available. The Speech Studio can help you segment long audio files and create transcriptions. The long-audio segmentation service will use the [Batch Transcription API](batch-transcription.md) feature of speech-to-text.

-> This service will be charged toward your speech-to-text subscription usage. The long-audio segmentation service is only supported with standard (S0) Speech resources.

-If you don't have transcriptions for your audio recordings, use the **Audio only** option to upload your data. Our system can help you segment and transcribe your audio files. Keep in mind, this service will be charged toward your speech-to-text subscription usage.

-> The long-audio segmentation service will leverage the batch transcription feature of speech-to-text, which only supports standard subscription (S0) users.

-Use the Speech Studio to [add a deployment endpoint](#add-a-deployment-endpoint) for your custom neural voice. You can use either the Speech Studio or text-to-speech REST API to [suspend or resume](#suspend-and-resume-an-endpoint) a custom neural voice endpoint.

-To use your custom neural voice, you must specify the voice model name, use the custom URI directly in an HTTP request, and use the same Speech resource to pass through the authentication of the text-to-speech service.

-The custom endpoint is functionally identical to the standard endpoint that's used for text-to-speech requests.

-One difference is that the `EndpointId` must be specified to use the custom voice via the Speech SDK. You can start with the [text-to-speech quickstart](get-started-text-to-speech.md) and then update the code with the `EndpointId` and `SpeechSynthesisVoiceName`.

-description: Learn how to get Speech service Speech-to-text Session ID and Transcription ID

-# How to get Speech-to-text Session ID and Transcription ID

-If you use [Speech-to-text](speech-to-text.md) and need to open a support case, you are often asked to provide a *Session ID* or *Transcription ID* of the problematic transcriptions to debug the issue. This article explains how to get these IDs.

-In case of [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) you need to "inject" the session information in the requests. See details in [this section](#provide-session-id-using-rest-api-for-short-audio).

-Unlike Speech SDK, [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) does not automatically generate a Session ID. You need to generate it yourself and provide it within the REST request.

-[Batch transcription API](batch-transcription.md) is a subset of the [Speech-to-text REST API](rest-speech-to-text.md).

-The custom neural voice lets you build higher-quality voice models while requiring less data. You can develop more realistic, natural, and conversational voices. Your customers and end users will benefit from the latest Text-to-Speech technology, in a responsible way.

-In this article, you'll learn how to evaluate pronunciation with the Speech-to-Text capability through the Speech SDK. To [get pronunciation assessment results](#get-pronunciation-assessment-results), you'll apply the `PronunciationAssessmentConfig` settings to a `SpeechRecognizer` object.

-> Usage of pronunciation assessment costs the same as standard Speech to Text pay-as-you-go [pricing](https://azure.microsoft.com/pricing/details/cognitive-services/speech-services). Pronunciation assessment doesn't yet support commitment tier pricing.

-Neural Text-to-SpeechΓÇ»(Neural TTS) turns input text or SSML (Speech Synthesis Markup Language) into lifelike synthesized speech. Speech audio output can be accompanied by viseme ID, Scalable Vector Graphics (SVG), or blend shapes. Using a 2D or 3D rendering engine, you can use these viseme events to animate your avatar.

-* [Try the text-to-speech quickstart](get-started-text-to-speech.md)

-description: The Speech Service SDK supports numerous programming languages for speech-to-text and text-to-speech conversion, along with speech translation. This article discusses memory management tooling built into the SDK.

-1. Go to **Real-time Speech-to-text** in [Speech Studio](https://aka.ms/speechstudio/speechtotexttool).

-> When you use conversational language understanding with the Speech SDK, you are charged both for the Speech-to-text recognition request and the Language service request for CLU. For more information about pricing for conversational language understanding, see [Language service pricing](https://azure.microsoft.com/pricing/details/cognitive-services/language-service/).

-There's no cost to use custom keyword to generate models, including both Basic and Advanced models. There's also no cost to run models on-device with the Speech SDK when used in conjunction with other Speech service features such as speech-to-text.

-Keyword verification is always used in combination with speech-to-text. There's no cost to use keyword verification beyond the cost of speech-to-text.

-### Keyword verification and speech-to-text

-When keyword verification is used, it's always in combination with speech-to-text. Both services run in parallel, which means audio is sent to both services for simultaneous processing.

-

-Running keyword verification and speech-to-text in parallel yields the following benefits:

-* **No other latency on speech-to-text results**: Parallel execution means that keyword verification adds no latency. The client receives speech-to-text results as quickly. If keyword verification determines the keyword wasn't present in the audio, speech-to-text processing is terminated. This action protects against unnecessary speech-to-text processing. Network and cloud model processing increases the user-perceived latency of voice activation. For more information, see [Recommendations and guidelines](keyword-recognition-guidelines.md).

-* **Forced keyword prefix in speech-to-text results**: Speech-to-text processing ensures that the results sent to the client are prefixed with the keyword. This behavior allows for increased accuracy in the speech-to-text results for speech that follows the keyword.

-* **Increased speech-to-text timeout**: Because of the expected presence of the keyword at the beginning of audio, speech-to-text allows for a longer pause of up to five seconds after the keyword before it determines the end of speech and terminates speech-to-text processing. This behavior ensures that the user experience is correctly handled for staged commands (*\<keyword> \<pause> \<command>*) and chained commands (*\<keyword> \<command>*).

-The Speech SDK enables seamless use of on-device models generated by using custom keyword with keyword verification and speech-to-text. It transparently handles:

-| End-to-end keyword recognition with speech-to-text | Best suited for products that will use a customized on-device keyword model from custom keyword with Azure Speech keyword verification and speech-to-text. This scenario is the most common. | <ul><li>[Voice assistant sample code](https://github.com/Azure-Samples/Cognitive-Services-Voice-Assistant)</li><li>[Tutorial: Voice enable your assistant built using Azure Bot Service with the C# Speech SDK](./tutorial-voice-enable-your-bot-speech-sdk.md)</li><li>[Tutorial: Create a custom commands application with simple voice commands](./how-to-develop-custom-commands-application.md)</li></ul> |

-* [Speech-to-text recognition](#speech-to-text) when you need to identify the language in an audio source and then transcribe it to text.

-Whether you use language identification with [speech-to-text](#speech-to-text) or with [speech translation](#speech-translation), there are some common concepts and configuration options.

-## Speech-to-text

-You use Speech-to-text recognition when you need to identify the language in an audio source and then transcribe it to text. For more information, see [Speech-to-text overview](speech-to-text.md).

-> Speech-to-text recognition with at-start language identification is supported with Speech SDKs in C#, C++, Python, Java, JavaScript, and Objective-C. Speech-to-text recognition with continuous language identification is only supported with Speech SDKs in C#, C++, Java, JavaScript, and Python.

-> Currently for speech-to-text recognition with continuous language identification, you must create a SpeechConfig from the `wss://{region}.stt.speech.microsoft.com/speech/universal/v2` endpoint string, as shown in code examples. In a future SDK release you won't need to set it.

-See more examples of speech-to-text recognition with language identification on [GitHub](https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/samples/csharp/sharedcontent/console/speech_recognition_with_language_id_samples.cs).

-See more examples of speech-to-text recognition with language identification on [GitHub](https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/samples/cpp/windows/console/samples/speech_recognition_samples.cpp).

-See more examples of speech-to-text recognition with language identification on [GitHub](https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/samples/java/jre/console/src/com/microsoft/cognitiveservices/speech/samples/console/SpeechRecognitionSamples.java).

-See more examples of speech-to-text recognition with language identification on [GitHub](https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/samples/python/console/speech_sample.py).

-### Speech-to-text custom models

-## Speech-to-text batch transcription

-## Speech-to-Text

-Azure [Speech-to-Text](speech-to-text.md) supports real-time language identification for multilingual language learning scenarios, help human-human interaction with better understanding and readable context.

-## Text-to-Speech

-[Text-to-Speech](text-to-speech.md) prebuilt neural voices can read out learning materials natively and empower self-served learning. A broad portfolio of [languages and voices](language-support.md?tabs=tts) are supported for AI teacher, content read aloud capabilities, and more. Microsoft is continuously working on bringing new languages to the world.

-* [What is Speech-to-Text](speech-to-text.md)

-* [What is Text-to-Speech](text-to-speech.md)

-description: The Speech service supports numerous languages for speech-to-text and text-to-speech conversion, along with speech translation. This article provides a comprehensive list of language support by service feature.

-The following tables summarize language support for [speech-to-text](speech-to-text.md), [text-to-speech](text-to-speech.md), [pronunciation assessment](how-to-pronunciation-assessment.md), [speech translation](speech-translation.md), [speaker recognition](speaker-recognition-overview.md), and additional service features.

-You can also get a list of locales and voices supported for each specific region or endpoint through the [Speech SDK](speech-sdk.md), [Speech-to-text REST API](rest-speech-to-text.md), [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) and [Text-to-speech REST API](rest-text-to-speech.md#get-a-list-of-voices).

-# [Speech-to-text](#tab/stt)

-The table in this section summarizes the locales and voices supported for Speech-to-text. Please see the table footnotes for more details.

-Additional remarks for Speech-to-text locales are included in the [Custom Speech](#custom-speech) section below.

-> Try out the [Real-time Speech-to-text tool](https://speech.microsoft.com/portal/speechtotexttool) without having to use any code.

-To improve Speech-to-text recognition accuracy, customization is available for some languages and base models. Depending on the locale, you can upload audio + human-labeled transcripts, plain text, structured text, and pronunciation data. By default, plain text customization is supported for all available base models. To learn more about customization, see [Custom Speech](./custom-speech-overview.md).

-# [Text-to-speech](#tab/tts)

-The tables in this section summarizes the locales and voices supported for Text-to-speech. Please see the table footnotes for more details.

-Additional remarks for Text-to-speech locales are included in the [Voice styles and roles](#voice-styles-and-roles), [Prebuilt neural voices](#prebuilt-neural-voices), and [Custom Neural Voice](#custom-neural-voice) sections below.

-The table in this section summarizes the locales supported for Pronunciation assessment, and each language is available on all [Speech-to-text regions](regions.md#speech-service).

-The table in this section summarizes the locales supported for Speech translation. Speech translation supports different languages for speech-to-speech and speech-to-text translation. The available target languages depend on whether the translation target is speech or text.

-To set the input speech recognition language, specify the full locale with a dash (`-`) separator. See the [speech-to-text language table](?tabs=stt#supported-languages). The default language is `en-US` if you don't specify a language.

-description: Learn how to use audio and transcription logging for speech-to-text and speech translation.

-You can enable logging for both audio input and recognized speech when using [speech-to-text](get-started-speech-to-text.md) or [speech translation](get-started-speech-to-text.md). For speech translation, only the audio and transcription of the original audio are logged. The translations aren't logged. This article describes how to enable, access and delete the audio and transcription logs.

-#### Enable logging for speech-to-text with the Speech SDK

-#### Enable logging for speech-to-text REST API for short audio

-If you use [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) and want to enable audio and transcription logging, you need to use the query parameter and value `storeAudio=true` as a part of your REST request. A sample request looks like this:

-To disable audio and transcription logging for a custom model endpoint, you must update the persistent endpoint logging setting using the [Speech-to-text REST API](rest-speech-to-text.md). There isn't a way to disable logging for an existing custom model endpoint using the Speech Studio.

-To turn off logging for a custom endpoint, use the [Endpoints_Update](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/operations/Endpoints_Update) operation of the [Speech-to-text REST API](rest-speech-to-text.md). Construct the request body according to the following instructions:

-You can access audio and transcription logs using [Speech-to-text REST API](#get-audio-and-transcription-logs-with-speech-to-text-rest-api). For [custom model](how-to-custom-speech-deploy-model.md) endpoints, you can also use [Speech Studio](#get-audio-and-transcription-logs-with-speech-studio). See details in the following sections.

-### Get audio and transcription logs with Speech-to-text REST API

-### Get log IDs with Speech-to-text REST API

-To delete audio and transcription logs you must use the [Speech-to-text REST API](rest-speech-to-text.md). There isn't a way to delete logs using the Speech Studio.

-For details about how to get Log IDs, see a previous section [Get log IDs with Speech-to-text REST API](#get-log-ids-with-speech-to-text-rest-api).

-* [Speech-to-text quickstart](get-started-speech-to-text.md)

-Batch synthesis API supports all [text-to-speech voices and styles](language-support.md?tabs=tts).

-Batch synthesis API supports all [text-to-speech audio output formats](rest-text-to-speech.md#audio-outputs).

-Compared to v2, the v3 version of the Speech services REST API for speech-to-text is more reliable, easier to use, and more consistent with APIs for similar services. Most teams can migrate from v2 to v3 in a day or two.

-> The Speech-to-text REST API v2.0 is deprecated and will be retired by February 29, 2024. Please migrate your applications to the Speech-to-text REST API v3.1. Complete the steps in this article and then see the [Migrate code from v3.0 to v3.1 of the REST API](migrate-v3-0-to-v3-1.md) guide for additional requirements.

-Endpoint host names have changed from `{region}.cris.ai` to `{region}.api.cognitive.microsoft.com`. Paths to the new endpoints no longer contain `api/` because it's part of the hostname. The [Speech-to-text REST API v3.0](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-0) reference documentation lists valid regions and paths.

-* [Speech-to-text REST API](rest-speech-to-text.md)

-* [Speech-to-text REST API v3.0 reference](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-0)

-The Speech-to-text REST API is used for [Batch transcription](batch-transcription.md) and [Custom Speech](custom-speech-overview.md). Changes from version 3.0 to 3.1 are described in the sections below.

-> Speech-to-text REST API v3.1 is generally available. Version 3.0 of the [Speech to Text REST API](rest-speech-to-text.md) will be retired.

-> Don't use Speech-to-text REST API v3.0 to retrieve a transcription created via Speech-to-text REST API v3.1. You'll see an error message such as the following: "The API version cannot be used to access this transcription. Please use API version v3.1 or higher."

-* [Speech-to-text REST API](rest-speech-to-text.md)

-* [Speech-to-text REST API v3.1 reference](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1)

-* [Speech-to-text REST API v3.0 reference](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-0)

-We're retiring two features from [Text-to-Speech](index-text-to-speech.yml) capabilities as detailed below.

-There are two kinds of languages: speech-to-text and text-only:

-| What the user can do in the conversation | Speech-to-text | Text-only |

-> For lists of available speech-to-text and text translation languages, see [supported languages](./language-support.md).

-description: The Speech service provides speech-to-text, text-to-speech, and speech translation capabilities with an Azure resource. Add speech to your applications, tools, and devices with the Speech SDK, Speech Studio, or REST APIs.

-The Speech service provides speech-to-text and text-to-speech capabilities with an [Azure Speech resource](~/articles/cognitive-services/cognitive-services-apis-create-account.md?tabs=speech#create-a-new-azure-cognitive-services-resource). You can transcribe speech to text with high accuracy, produce natural-sounding text-to-speech voices, translate spoken audio, and use speaker recognition during conversations.

-### Speech-to-text

-Use [speech-to-text](speech-to-text.md) to transcribe audio into text, either in [real-time](#real-time-speech-to-text) or asynchronously with [batch transcription](#batch-transcription).

-> You can try real-time speech-to-text in [Speech Studio](https://aka.ms/speechstudio/speechtotexttool) without signing up or writing any code.

-### Real-time speech-to-text

-With [real-time speech-to-text](get-started-speech-to-text.md), the audio is transcribed as speech is recognized from a microphone or file. Use real-time speech-to-text for applications that need to transcribe audio in real-time such as:

-### Text-to-speech

-[Speech translation](speech-translation.md) enables real-time, multilingual translation of speech to your applications, tools, and devices. Use this feature for speech-to-speech and speech-to-text translation.

-[Language identification](language-identification.md) is used to identify languages spoken in audio when compared against a list of [supported languages](language-support.md). Use language identification by itself, with speech-to-text recognition, or with speech translation.

-[Intent recognition](./intent-recognition.md): Use speech-to-text with conversational language understanding to derive user intents from transcribed speech and act on voice commands.

-* [Speech-to-text quickstart](get-started-speech-to-text.md)

-* [Text-to-speech quickstart](get-started-text-to-speech.md)

-* [Get started with speech-to-text](get-started-speech-to-text.md)

-* [Get started with text-to-speech](get-started-text-to-speech.md)

-This article describes how to use [Power Automate](/power-automate/getting-started) and the [Azure Cognitive Services for Batch Speech-to-text connector](/connectors/cognitiveservicesspe/) to transcribe audio files from an Azure Storage container. The connector uses the [Batch Transcription REST API](batch-transcription.md), but you don't need to write any code to use it. If the connector doesn't meet your requirements, you can still use the [REST API](rest-speech-to-text.md#transcriptions) directly.

-In addition to [Power Automate](/power-automate/getting-started), you can use the [Azure Cognitive Services for Batch Speech-to-text connector](/connectors/cognitiveservicesspe/) with [Power Apps](/power-apps) and [Logic Apps](../../logic-apps/index.yml).

-1. Select **+ New step** to begin adding a new operation for the [Azure Cognitive Services for Batch Speech-to-text connector](/connectors/cognitiveservicesspe/).

-1. Enter "batch speech-to-text" in the search connectors and actions box to narrow the results.

-1. Select the **Azure Cognitive Services for Batch Speech-to-text** connector.

- > For more information about create transcription parameters, see the [Azure Cognitive Services for Batch Speech-to-text](/connectors/cognitiveservicesspe/#create-transcription) documentation.

-Pronunciation assessment uses the Speech-to-Text capability to provide subjective and objective feedback for language learners. Practicing pronunciation and getting timely feedback are essential for improving language skills. Assessments driven by experienced teachers can take a lot of time and effort and makes a high-quality assessment expensive for learners. Pronunciation assessment can help make the language assessment more engaging and accessible to learners of all backgrounds.

-> Usage of pronunciation assessment costs the same as standard Speech to Text pay-as-you-go [pricing](https://azure.microsoft.com/pricing/details/cognitive-services/speech-services). Pronunciation assessment doesn't yet support commitment tier pricing.

-* [Speech-to-text quickstart](../get-started-speech-to-text.md)

-* [Text-to-speech quickstart](../get-started-text-to-speech.md)

- - Balanced coverage for pronunciations. Include all letters from A to Z so the Text-to-Speech engine learns how to pronounce each letter in your style.

-description: A list of available regions and endpoints for the Speech service, including speech-to-text, text-to-speech, and speech translation.

-The following regions are supported for Speech service features such as speech-to-text, text-to-speech, pronunciation assessment, and translation. The geographies are listed in alphabetical order.

-* Speech-to-text and text-to-speech container versions were updated in March 2023.

-* Custom Speech-to-Text container disconnected mode was released in January 2023.

-* Text-to-speech Batch synthesis API is available in public preview.

-* Speech-to-text REST API version 3.1 is generally available.

-# [Text-to-speech service](#tab/text-to-speech)

-# [Speech-to-text service](#tab/speech-to-text)

- 2. For persistence redirect traffic to the new STS token service and Speech service endpoint. (For Text-to-Speech, reference sample code: [GitHub: TTS public voice switching region](https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/samples/csharp/sharedcontent/console/speech_synthesis_samples.cs#L880).

-description: Learn how to use Speech-to-text REST API for short audio to convert speech to text.

-# Speech-to-text REST API for short audio

-Use cases for the speech-to-text REST API for short audio are limited. Use it only in cases where you can't use the [Speech SDK](speech-sdk.md).

-Before you use the speech-to-text REST API for short audio, consider the following limitations:

-* [Batch transcription](batch-transcription.md) and [Custom Speech](custom-speech-overview.md) are not supported via REST API for short audio. You should always use the [Speech to Text REST API](rest-speech-to-text.md) for batch transcription and Custom Speech.

-Before you use the speech-to-text REST API for short audio, understand that you need to complete a token exchange as part of authentication to access the service. For more information, see [Authentication](#authentication).

-This table lists required and optional headers for speech-to-text requests:

-description: Get reference documentation for Speech-to-text REST API.

-# Speech-to-text REST API

-Speech-to-text REST API is used for [Batch transcription](batch-transcription.md) and [Custom Speech](custom-speech-overview.md).

-> Speech-to-text REST API v3.1 is generally available. Version 3.0 of the [Speech to Text REST API](rest-speech-to-text.md) will be retired. For more information, see the [Migrate code from v3.0 to v3.1 of the REST API](migrate-v3-0-to-v3-1.md) guide.

-> [See the Speech to Text API v3.1 reference documentation](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-1/)

-> [See the Speech to Text API v3.0 reference documentation](https://eastus.dev.cognitive.microsoft.com/docs/services/speech-to-text-api-v3-0/)

-Use Speech-to-text REST API to:

-Speech-to-text REST API includes such features as:

-This table includes all the web hook operations that are available with the speech-to-text REST API.

-# Text-to-speech REST API

-> Use cases for the text-to-speech REST API are limited. Use it only in cases where you can't use the [Speech SDK](speech-sdk.md). For example, with the Speech SDK you can [subscribe to events](how-to-speech-synthesis.md#subscribe-to-synthesizer-events) for more insights about the text-to-speech processing and results.

-The text-to-speech REST API supports neural text-to-speech voices, which support specific languages and dialects that are identified by locale. Each available endpoint is associated with a region. A Speech resource key for the endpoint or region that you plan to use is required. Here are links to more information:

-Before you use the text-to-speech REST API, understand that you need to complete a token exchange as part of authentication to access the service. For more information, see [Authentication](#authentication).

-This table lists required and optional headers for text-to-speech requests:

-These regions are supported for text-to-speech through the REST API. Be sure to select the endpoint that matches your Speech resource region.

-This table lists required and optional headers for text-to-speech requests:

-If you're using a custom neural voice, the body of a request can be sent as plain text (ASCII or UTF-8). Otherwise, the body of each `POST` request is sent as [SSML](speech-synthesis-markup.md). SSML allows you to choose the voice and language of the synthesized speech that the text-to-speech feature returns. For a complete list of supported voices, see [Language and voice support for the Speech service](language-support.md?tabs=tts).

- Microsoft Speech Service Text-to-Speech API

- - Speech-to-text

- - Text-to-speech

-This section contains Speech Services endpoint information for the usage with [Speech SDK](speech-sdk.md), [Speech-to-text REST API](rest-speech-to-text.md), and [Text-to-speech REST API](rest-text-to-speech.md).

-| [Speech-to-text REST API](rest-speech-to-text.md) | `https://<REGION_IDENTIFIER>.api.cognitive.microsoft.us/<URL_PATH>` |

-| [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) | `https://<REGION_IDENTIFIER>.stt.speech.azure.us/<URL_PATH>` |

-| [Text-to-speech REST API](rest-text-to-speech.md) | `https://<REGION_IDENTIFIER>.tts.speech.azure.us/<URL_PATH>` |

-| Speech-to-text | `wss://usgovarizona.stt.speech.azure.us` |

-| Text-to-Speech | `https://usgovarizona.tts.speech.azure.us` |

-| Speech-to-text | `wss://usgovvirginia.stt.speech.azure.us` |

-| Text-to-Speech | `https://usgovvirginia.tts.speech.azure.us` |

- - Speech-to-text

- - Text-to-speech

-This section contains Speech Services endpoint information for the usage with [Speech SDK](speech-sdk.md), [Speech-to-text REST API](rest-speech-to-text.md), and [Text-to-speech REST API](rest-text-to-speech.md).

-| [Speech-to-text REST API](rest-speech-to-text.md) | `https://<REGION_IDENTIFIER>.api.cognitive.azure.cn/<URL_PATH>` |

-| [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) | `https://<REGION_IDENTIFIER>.stt.speech.azure.cn/<URL_PATH>` |

-| [Text-to-speech REST API](rest-text-to-speech.md) | `https://<REGION_IDENTIFIER>.tts.speech.azure.cn/<URL_PATH>` |

-| Speech-to-text | `wss://chinaeast2.stt.speech.azure.cn` |

-| Text-to-Speech | `https://chinaeast2.tts.speech.azure.cn` |

-| Speech-to-text | `wss://chinanorth2.stt.speech.azure.cn` |

-| Text-to-Speech | `https://chinanorth2.tts.speech.azure.cn` |

-> The volume mount settings are only applicable for [Custom Speech-to-text](speech-container-cstt.md) containers.

-description: Install and run custom speech-to-text containers with Docker to perform speech recognition, transcription, generation, and more on-premises.

-# Custom speech-to-text containers with Docker

-The Custom speech-to-text container transcribes real-time speech or batch audio recordings with intermediate results. You can use a custom model that you created in the [Custom Speech portal](https://speech.microsoft.com/customspeech). In this article, you'll learn how to download, install, and run a Custom speech-to-text container.

-The Custom speech-to-text container image for all supported versions and locales can be found on the [Microsoft Container Registry (MCR)](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/custom-speech-to-text/tags) syndicate. It resides within the `azure-cognitive-services/speechservices/` repository and is named `custom-speech-to-text`.

-> The `locale` and `voice` for custom speech-to-text containers is determined by the custom model ingested by the container.

-Before you [run](#run-the-container-with-docker-run) the container, you can optionally get the available display models information and choose to download those models into your speech-to-text container to get highly improved final display output. Display model download is available with custom-speech-to-text container version 3.1.0 and later.

-In order to prepare and configure a disconnected custom speech-to-text container you will need two separate speech resources:

-The Custom Speech-to-Text container provides a default directory for writing the license file and billing log at runtime. The default directories are /license and /output respectively.

-[Try the speech-to-text quickstart](get-started-speech-to-text.md) using host authentication instead of key and region.

-description: Using Kubernetes and Helm to define the speech-to-text and text-to-speech container images, we'll create a Kubernetes package. This package will be deployed to a Kubernetes cluster on-premises.

-One option to manage your Speech containers on-premises is to use Kubernetes and Helm. Using Kubernetes and Helm to define the speech-to-text and text-to-speech container images, we'll create a Kubernetes package. This package will be deployed to a Kubernetes cluster on-premises. Finally, we'll explore how to test the deployed services and various configuration options. For more information about running Docker containers without Kubernetes orchestration, see [install and run Speech service containers](speech-container-howto.md).

-| **Speech-to-Text** | one decoder requires a minimum of 1,150 millicores. If the `optimizedForAudioFile` is enabled, then 1,950 millicores are required. (default: two decoders) | Required: 2 GB<br>Limited: 4 GB |

-| **Text-to-Speech** | one concurrent request requires a minimum of 500 millicores. If the `optimizeForTurboMode` is enabled, then 1,000 millicores are required. (default: two concurrent requests) | Required: 1 GB<br> Limited: 2 GB |

-# speech-to-text configurations

-# text-to-speech configurations

-The provided *Helm charts* pull the docker images of the Speech service, both text-to-speech and the speech-to-text services from the `mcr.microsoft.com` container registry.

-The installed Helm charts define *Helm tests*, which serve as a convenience for verification. These tests validate service readiness. To verify both **speech-to-text** and **text-to-speech** services, we'll execute the [Helm test][helm-test] command.

-RUNNING: speech-to-text-readiness-test

-PASSED: speech-to-text-readiness-test

-RUNNING: text-to-speech-readiness-test

-PASSED: text-to-speech-readiness-test

-| Speech-to-text | 4 core, 4-GB memory | 8 core, 8-GB memory |+4 to 8 GB memory|

-| Custom speech-to-text | 4 core, 4-GB memory | 8 core, 8-GB memory |+4 to 8 GB memory|

-| Neural text-to-speech | 6 core, 12-GB memory | 8 core, 16-GB memory |n/a|

-> For example, speech-to-text containers memory map portions of a large language model. We recommend that the entire file should fit in memory. You need to add an additional 4 to 8 GB to load the speech models (see above table).

-| WS | `ws://localhost:5000` | [Speech-to-text](speech-container-stt.md#use-the-container)<br/><br/>[Custom speech-to-text](speech-container-cstt.md#use-the-container) |

-| HTTP | `http://localhost:5000` | [Neural text-to-speech](speech-container-ntts.md#use-the-container)<br/><br/>[Speech language identification](speech-container-lid.md#use-the-container) |

-> To get the most useful results, use the Speech language identification container with the [speech-to-text](speech-container-stt.md) or [custom speech-to-text](speech-container-cstt.md) containers.

-## Run with the speech-to-text container

-If you want to run the language identification container with the [speech-to-text](speech-container-stt.md) container, you can use this [docker image](https://hub.docker.com/r/antsu/on-prem-client). After both containers have been started, use this `docker run` command to execute `speech-to-text-with-languagedetection-client`:

-description: Install and run neural text-to-speech containers with Docker to perform speech synthesis and more on-premises.

-# Text-to-speech containers with Docker

-The neural text-to-speech container converts text to natural-sounding speech by using deep neural network technology, which allows for more natural synthesized speech.. In this article, you'll learn how to download, install, and run a Text-to-speech container.

-The neural text-to-speech container image for all supported versions and locales can be found on the [Microsoft Container Registry (MCR)](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/neural-text-to-speech/tags) syndicate. It resides within the `azure-cognitive-services/speechservices/` repository and is named `neural-text-to-speech`.

-> We retired the standard speech synthesis voices and standard [text-to-speech](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/text-to-speech/tags) container on August 31, 2021. You should use neural voices with the [neural-text-to-speech](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/neural-text-to-speech/tags) container instead. For more information on updating your application, see [Migrate from standard voice to prebuilt neural voice](./how-to-migrate-to-prebuilt-neural-voice.md).

-> The `latest` tag pulls the `en-US` locale and `en-us-arianeural` voice. For additional locales and voices, see [text-to-speech container images](#container-images).

-When you run the text-to-speech container, configure the port, memory, and CPU according to the text-to-speech container [requirements and recommendations](speech-container-howto.md#container-requirements-and-recommendations).

-* Runs a neural text-to-speech container from the container image.

-[Try the text-to-speech quickstart](get-started-text-to-speech.md) using host authentication instead of key and region.

-When you construct a neural text-to-speech HTTP POST, the [SSML](speech-synthesis-markup.md) message requires a `voice` element with a `name` attribute. The [locale of the voice](language-support.md?tabs=tts) must correspond to the locale of the container model.

-| [Speech-to-text](speech-container-stt.md) | Transcribes continuous real-time speech or batch audio recordings with intermediate results. | Latest: 3.12.0<br/><br/>For all supported versions and locales, see the [Microsoft Container Registry (MCR)](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/speech-to-text/tags) and [JSON tags](https://mcr.microsoft.com/v2/azure-cognitive-services/speechservices/speech-to-text/tags/list).|

-| [Custom speech-to-text](speech-container-cstt.md) | Using a custom model from the [Custom Speech portal](https://speech.microsoft.com/customspeech), transcribes continuous real-time speech or batch audio recordings into text with intermediate results. | Latest: 3.12.0<br/><br/>For all supported versions and locales, see the [Microsoft Container Registry (MCR)](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/custom-speech-to-text/tags) and [JSON tags](https://mcr.microsoft.com/v2/azure-cognitive-services/speechservices/speech-to-text/tags/list). |

-| [Neural text-to-speech](speech-container-ntts.md) | Converts text to natural-sounding speech by using deep neural network technology, which allows for more natural synthesized speech. | Latest: 2.11.0<br/><br/>For all supported versions and locales, see the [Microsoft Container Registry (MCR)](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/neural-text-to-speech/tags) and [JSON tags](https://mcr.microsoft.com/v2/azure-cognitive-services/speechservices/neural-text-to-speech/tags/list). |

-description: Install and run speech-to-text containers with Docker to perform speech recognition, transcription, generation, and more on-premises.

-# Speech-to-text containers with Docker

-The Speech-to-text container transcribes real-time speech or batch audio recordings with intermediate results. In this article, you'll learn how to download, install, and run a Speech-to-text container.

-The Speech-to-text container image for all supported versions and locales can be found on the [Microsoft Container Registry (MCR)](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/speech-to-text/tags) syndicate. It resides within the `azure-cognitive-services/speechservices/` repository and is named `speech-to-text`.

-> The `latest` tag pulls the latest image for the `en-US` locale. For additional versions and locales, see [speech-to-text container images](#container-images).

-When you run the speech-to-text container, configure the port, memory, and CPU according to the speech-to-text container [requirements and recommendations](speech-container-howto.md#container-requirements-and-recommendations).

-[Try the speech-to-text quickstart](get-started-speech-to-text.md) using host authentication instead of key and region.

-In some cases, you can't or shouldn't use the [Speech SDK](speech-sdk.md). In those cases, you can use REST APIs to access the Speech service. For example, use the [Speech-to-text REST API](rest-speech-to-text.md) for [batch transcription](batch-transcription.md) and [custom speech](custom-speech-overview.md).

-The following video shows how to install the [Speech SDK for C#](quickstarts/setup-platform.md) and write a simple .NET console application for speech-to-text.

-* [Speech-to-text REST API](rest-speech-to-text.md)

-* [Text-to-speech REST API](rest-text-to-speech.md)

-Speech service has REST APIs for [Speech-to-text](rest-speech-to-text.md) and [Text-to-speech](rest-text-to-speech.md). Consider the following information for the private-endpoint-enabled scenario.

-Speech-to-text has two REST APIs. Each API serves a different purpose, uses different endpoints, and requires a different approach when you're using it in the private-endpoint-enabled scenario.

-The Speech-to-text REST APIs are:

-Usage of the Speech-to-text REST API for short audio and the Text-to-speech REST API in the private endpoint scenario is the same. It's equivalent to the [Speech SDK case](#speech-resource-with-a-custom-domain-name-and-a-private-endpoint-usage-with-the-speech-sdk) described later in this article.

-Speech-to-text REST API uses a different set of endpoints, so it requires a different approach for the private-endpoint-enabled scenario.

-#### Speech-to-text REST API

-Usually, Speech resources use [Cognitive Services regional endpoints](../cognitive-services-custom-subdomains.md#is-there-a-list-of-regional-endpoints) for communicating with the [Speech-to-text REST API](rest-speech-to-text.md). These resources have the following naming format: <p/>`{region}.api.cognitive.microsoft.com`.

-#### Speech-to-text REST API for short audio and Text-to-speech REST API

-The [Speech-to-text REST API for short audio](rest-speech-to-text-short.md) and the [Text-to-speech REST API](rest-text-to-speech.md) use two types of endpoints:

-The detailed description of the special endpoints and how their URL should be transformed for a private-endpoint-enabled Speech resource is provided in [this subsection](#construct-endpoint-url) about usage with the Speech SDK. The same principle described for the SDK applies for the Speech-to-text REST API for short audio and the Text-to-speech REST API.

-Get familiar with the material in the subsection mentioned in the previous paragraph and see the following example. The example describes the Text-to-speech REST API. Usage of the Speech-to-text REST API for short audio is fully equivalent.

-> When you're using the Speech-to-text REST API for short audio and Text-to-speech REST API in private endpoint scenarios, use a resource key passed through the `Ocp-Apim-Subscription-Key` header. (See details for [Speech-to-text REST API for short audio](rest-speech-to-text-short.md#request-headers) and [Text-to-speech REST API](rest-text-to-speech.md#request-headers))

-**Text-to-speech REST API usage example**

-See more details in the [Text-to-speech REST API documentation](rest-text-to-speech.md).

-Usually in SDK scenarios (as well as in the Speech-to-text REST API for short audio and Text-to-speech REST API scenarios), Speech resources use the dedicated regional endpoints for different service offerings. The DNS name format for these endpoints is:

-| `stt` | [Speech-to-text](speech-to-text.md) |

-| `tts` | [Text-to-speech](text-to-speech.md) |

-So the earlier example (`westeurope.stt.speech.microsoft.com`) stands for a Speech-to-text endpoint in West Europe.

-#### Speech-to-text REST API

-Speech-to-text REST API usage is fully equivalent to the case of [private-endpoint-enabled Speech resources](#speech-to-text-rest-api).

-#### Speech-to-text REST API for short audio and Text-to-speech REST API

-In this case, usage of the Speech-to-text REST API for short audio and usage of the Text-to-speech REST API have no differences from the general case, with one exception. (See the following note.) You should use both APIs as described in the [speech-to-text REST API for short audio](rest-speech-to-text-short.md) and [Text-to-speech REST API](rest-text-to-speech.md) documentation.

-> When you're using the Speech-to-text REST API for short audio and Text-to-speech REST API in custom domain scenarios, use a Speech resource key passed through the `Ocp-Apim-Subscription-Key` header. (See details for [Speech-to-text REST API for short audio](rest-speech-to-text-short.md#request-headers) and [Text-to-speech REST API](rest-text-to-speech.md#request-headers))

-* [Speech-to-text REST API](rest-speech-to-text.md)

-* [Text-to-speech REST API](rest-text-to-speech.md)

-### Speech-to-text quotas and limits per resource

-This section describes speech-to-text quotas and limits per Speech resource. Unless otherwise specified, the limits aren't adjustable.

-You can use real-time speech-to-text with the [Speech SDK](speech-sdk.md) or the [speech-to-text REST API for short audio](rest-speech-to-text-short.md).

-> These limits apply to concurrent real-time speech-to-text requests and speech translation requests combined. For example, if you have 60 concurrent speech-to-text requests and 40 concurrent speech translation requests, you'll reach the limit of 100 concurrent requests.

-| [Speech-to-text REST API](rest-speech-to-text.md) limit | Not available for F0 | 300 requests per minute |

-### Text-to-speech quotas and limits per resource

-This section describes text-to-speech quotas and limits per Speech resource. Unless otherwise specified, the limits aren't adjustable.

-#### Common text-to-speech quotas and limits

-### Speech-to-text: increase real-time speech-to-text concurrent request limit

-By default, the number of concurrent real-time speech-to-text and speech translation [requests combined](#real-time-speech-to-text-and-speech-translation) is limited to 100 per resource in the base model, and 100 per custom endpoint in the custom model. For the standard pricing tier, you can increase this amount. Before submitting the request, ensure that you're familiar with the material discussed earlier in this article, such as the best practices to mitigate throttling.

-1. In **Summary**, describe what you want (for example, "Increase speech-to-text concurrency request limit").

- - A note that the request is about the speech-to-text quota.

-### Text-to-speech: increase concurrent request limit

-1. In **Summary**, describe what you want (for example, "Increase text-to-speech concurrency request limit").

- - A note that the request is about the text-to-speech quota.

-Phonetic alphabets are used with the [Speech Synthesis Markup Language (SSML)](speech-synthesis-markup.md) to improve the pronunciation of text-to-speech voices. To learn when and how to use each alphabet, see [Use phonemes to improve pronunciation](speech-synthesis-markup-pronunciation.md#phoneme-element).

-> You can try speech-to-text and text-to-speech in [Speech Studio](https://aka.ms/speechstudio/) without signing up or writing any code.

-* [Real-time speech-to-text](https://aka.ms/speechstudio/speechtotexttool): Quickly test speech-to-text by dragging audio files here without having to use any code. This is a demo tool for seeing how speech-to-text works on your audio samples. To explore the full functionality, see [What is speech-to-text?](speech-to-text.md).

-* [Custom Voice](https://aka.ms/speechstudio/customvoice): Create custom, one-of-a-kind voices for text-to-speech. You supply audio files and create matching transcriptions in Speech Studio, and then use the custom voices in your applications. To create and use custom voices via endpoints, see [Create and use your voice model](how-to-custom-voice-create-voice.md).

-* [Audio Content Creation](https://aka.ms/speechstudio/audiocontentcreation): A no-code approach for text-to-speech synthesis. You can use the output audio as-is, or as a starting point for further customization. You can build highly natural audio content for a variety of scenarios, such as audiobooks, news broadcasts, video narrations, and chat bots. For more information, see the [Audio Content Creation](how-to-audio-content-creation.md) documentation.

-You can use Speech Synthesis Markup Language (SSML) with text-to-speech to specify how the speech is pronounced. For example, you can use SSML with phonemes and a custom lexicon to improve pronunciation. You can also use SSML to define how a word or mathematical expression is pronounced.

-| `ph` | A string containing phones that specify the pronunciation of the word in the `phoneme` element. If the specified string contains unrecognized phones, text-to-speech rejects the entire SSML document and produces none of the speech output specified in the document.<br/><br/>For `ipa`, to stress one syllable by placing stress symbol before this syllable, you need to mark all syllables for the word. Or else, the syllable before this stress symbol will be stressed. For `sapi`, if you want to stress one syllable, you need to place the stress symbol after this syllable, whether or not all syllables of the word are marked.| Required |

-> The `lexicon` element is not supported by the [Long Audio API](migrate-to-batch-synthesis.md#text-inputs). For long-form text-to-speech, use the [batch synthesis API](batch-synthesis.md) (Preview) instead.

-The text-to-speech output for this example is "a squared plus b squared equals c squared".

-The Speech Synthesis Markup Language (SSML) with input text determines the structure, content, and other characteristics of the text-to-speech output. For example, you can use SSML to define a paragraph, a sentence, a break or a pause, or silence. You can wrap text with event tags such as bookmark or viseme that can be processed later by your application.

- Welcome <break /> to text-to-speech.

- Welcome <break strength="medium" /> to text-to-speech.

- Welcome <break time="750ms" /> to text-to-speech.

-Use Speech Synthesis Markup Language (SSML) to specify the text-to-speech voice, language, name, style, and role. You can use multiple voices in a single SSML document. Adjust the emphasis, speaking rate, pitch, and volume. You can also use SSML to insert pre-recorded audio, such as a sound effect or a musical note.

-At least one `voice` element must be specified within each SSML [speak](speech-synthesis-markup-structure.md#speak-root-element) element. This element determines the voice that's used for text-to-speech.

-| `name` | The voice used for text-to-speech output. For a complete list of supported prebuilt voices, see [Language support](language-support.md?tabs=tts).| Required|

-Within the `speak` element, you can specify multiple voices for text-to-speech output. These voices can be in different languages. For each voice, the text must be wrapped in a `voice` element.

-Within the `speak` element, you can specify multiple languages including `en-US` for text-to-speech output. For each adjusted language, the text must match the language and be wrapped in a `voice` element. This SSML snippet shows how to use `<lang xml:lang>` to change the speaking languages to `es-MX`, `en-US`, and `fr-FR`.

-The `prosody` element is used to specify changes to pitch, contour, range, rate, and volume for the text-to-speech output. The `prosody` element can contain text and the following elements: `audio`, `break`, `p`, `phoneme`, `prosody`, `say-as`, `sub`, and `s`.

-Because prosodic attribute values can vary over a wide range, the speech recognizer interprets the assigned values as a suggestion of what the actual prosodic values of the selected voice should be. Text-to-speech limits or substitutes values that aren't supported. Examples of unsupported values are a pitch of 1 MHz or a volume of 120.

- Enjoy using text-to-speech.

- Enjoy using text-to-speech.

- Welcome to <prosody pitch="high">Enjoy using text-to-speech.</prosody>

-> The `audio` element is not supported by the [Long Audio API](migrate-to-batch-synthesis.md#text-inputs). For long-form text-to-speech, use the [batch synthesis API](batch-synthesis.md) (Preview) instead.

-| `src` | The URI location of the audio file. The audio must be hosted on an internet-accessible HTTPS endpoint. HTTPS is required, and the domain hosting the file must present a valid, trusted TLS/SSL certificate. We recommend that you put the audio file into Blob Storage in the same Azure region as the text-to-speech endpoint to minimize the latency. | Required |

-You can use the `mstts:backgroundaudio` element to add background audio to your SSML documents or mix an audio file with text-to-speech. With `mstts:backgroundaudio`, you can loop an audio file in the background, fade in at the beginning of text-to-speech, and fade out at the end of text-to-speech.

-If the background audio provided is shorter than the text-to-speech or the fade out, it loops. If it's longer than the text-to-speech, it stops when the fade out has finished.

-> The `mstts:backgroundaudio` element is not supported by the [Long Audio API](migrate-to-batch-synthesis.md#text-inputs). For long-form text-to-speech, use the [batch synthesis API](batch-synthesis.md) (Preview) instead.

-description: Use the Speech Synthesis Markup Language to control pronunciation and prosody in text-to-speech.

-Speech Synthesis Markup Language (SSML) is an XML-based markup language that can be used to fine-tune the text-to-speech output attributes such as pitch, pronunciation, speaking rate, volume, and more. You have more control and flexibility compared to plain text input.

-> You're billed for each character that's converted to speech, including punctuation. Although the SSML document itself is not billable, optional elements that are used to adjust how the text is converted to speech, like phonemes and pitch, are counted as billable characters. For more information, see [text-to-speech pricing notes](text-to-speech.md#pricing-note).

-description: Get an overview of the benefits and capabilities of the speech-to-text feature of the Speech Service.

-# What is speech-to-text?

-In this overview, you learn about the benefits and capabilities of the speech-to-text feature of the Speech service, which is part of Azure Cognitive Services. Speech-to-text can be used for [real-time](#real-time-speech-to-text) or [batch transcription](#batch-transcription) of audio streams into text.

-For a full list of available speech-to-text languages, see [Language and voice support](language-support.md?tabs=stt).

-## Real-time speech-to-text

-With real-time speech-to-text, the audio is transcribed as speech is recognized from a microphone or file. Use real-time speech-to-text for applications that need to transcribe audio in real-time such as:

-With [Custom Speech](./custom-speech-overview.md), you can evaluate and improve the accuracy of speech recognition for your applications and products. A custom speech model can be used for [real-time speech-to-text](speech-to-text.md), [speech translation](speech-translation.md), and [batch transcription](batch-transcription.md).

-A custom model can be used to augment the base model to improve recognition of domain-specific vocabulary specific to the application by providing text data to train the model. It can also be used to improve recognition based for the specific audio conditions of the application by providing audio data with reference transcriptions. For more information, see [Custom Speech](./custom-speech-overview.md) and [Speech-to-text REST API](rest-speech-to-text.md).

-In this article, you learn about the benefits and capabilities of the speech translation service, which enables real-time, multi-language speech-to-speech and speech-to-text translation of audio streams.

-* Speech-to-text translation with recognition results.

-You'll find [Speech SDK speech-to-text and translation samples](https://github.com/Azure-Samples/cognitive-services-speech-sdk) on GitHub. These samples cover common scenarios, such as reading audio from a file or stream, continuous and single-shot recognition and translation, and working with custom models.

-description: In this Azure Speech CLI quickstart, you interact with speech-to-text, text-to-speech, and speech translation without having to write code.

-In this article, you'll learn how to use the Azure Speech CLI (also called SPX) to access Speech services such as speech-to-text, text-to-speech, and speech translation, without having to write any code. The Speech CLI is production ready, and you can use it to automate simple workflows in the Speech service by using `.bat` or shell scripts.

-## Speech-to-text (speech recognition)

-## Text-to-speech (speech synthesis)

-## Speech-to-text translation

-With the Speech CLI, you can also do speech-to-text translation. Run the following command to capture audio from your default microphone and output the translation as text. Keep in mind that you need to supply the `source` and `target` language with the `translate` command.

-The easiest way to run batch text-to-speech is to create a new `.tsv` (tab-separated-value) file, and use the `--foreach` command in the Speech CLI. You can create a `.tsv` file using your favorite text editor, for this example, let's call it `text_synthesis.tsv`:

-The Speech CLI is a command-line tool for using Speech service without having to write any code. The Speech CLI requires minimal setup. You can easily use it to experiment with key features of Speech service and see how it works with your use cases. Within minutes, you can run simple test workflows, such as batch speech-recognition from a directory of files or text-to-speech on a collection of strings from a file. Beyond simple workflows, the Speech CLI is production-ready, and you can scale it up to run larger processes by using automated `.bat` or shell scripts.

-To get started with the Speech CLI, see the [quickstart](spx-basics.md). This article shows you how to run some basic commands. It also gives you slightly more advanced commands for running batch operations for speech-to-text and text-to-speech. After you've read the basics article, you should understand the syntax well enough to start writing some custom commands or automate simple Speech service operations.

-# Generate a REST API client library for the Speech-to-text REST API

-> Speech service has several REST APIs for [Speech-to-text](rest-speech-to-text.md) and [Text-to-speech](rest-text-to-speech.md).

-> However only [Speech-to-text REST API](rest-speech-to-text.md) is documented in the Swagger specification. See the documents referenced in the previous paragraph for the information on all other Speech Services REST APIs.

-* [Speech-to-text REST API](rest-speech-to-text.md)

-description: Get an overview of the benefits and capabilities of the text-to-speech feature of the Speech service.

-# What is text-to-speech?

-In this overview, you learn about the benefits and capabilities of the text-to-speech feature of the Speech service, which is part of Azure Cognitive Services.

-Text-to-speech enables your applications, tools, or devices to convert text into humanlike synthesized speech. The text-to-speech capability is also known as speech synthesis. Use humanlike prebuilt neural voices out of the box, or create a custom neural voice that's unique to your product or brand. For a full list of supported voices, languages, and locales, see [Language and voice support for the Speech service](language-support.md?tabs=tts).

-Text-to-speech includes the following features:

-### More about neural text-to-speech features

-The text-to-speech feature of the Speech service on Azure has been fully upgraded to the neural text-to-speech engine. This engine uses deep neural networks to make the voices of computers nearly indistinguishable from the recordings of people. With the clear articulation of words, neural text-to-speech significantly reduces listening fatigue when users interact with AI systems.

-The patterns of stress and intonation in spoken language are called _prosody_. Traditional text-to-speech systems break down prosody into separate linguistic analysis and acoustic prediction steps that are governed by independent models. That can result in muffled, buzzy voice synthesis.

-Here's more information about neural text-to-speech features in the Speech service, and how they overcome the limits of traditional text-to-speech systems:

-* **Real-time speech synthesis**: Use the [Speech SDK](./get-started-text-to-speech.md) or [REST API](rest-text-to-speech.md) to convert text-to-speech by using [prebuilt neural voices](language-support.md?tabs=tts) or [custom neural voices](custom-neural-voice.md).

-* **Asynchronous synthesis of long audio**: Use the [batch synthesis API](batch-synthesis.md) (Preview) to asynchronously synthesize text-to-speech files longer than 10 minutes (for example, audio books or lectures). Unlike synthesis performed via the Speech SDK or speech-to-text REST API, responses aren't returned in real-time. The expectation is that requests are sent asynchronously, responses are polled for, and synthesized audio is downloaded when the service makes it available.

-* **Prebuilt neural voices**: Microsoft neural text-to-speech capability uses deep neural networks to overcome the limits of traditional speech synthesis with regard to stress and intonation in spoken language. Prosody prediction and voice synthesis happen simultaneously, which results in more fluid and natural-sounding outputs. Each prebuilt neural voice model is available at 24kHz and high-fidelity 48kHz. You can use neural voices to:

-* **Fine-tuning text-to-speech output with SSML**: Speech Synthesis Markup Language (SSML) is an XML-based markup language that's used to customize text-to-speech outputs. With SSML, you can adjust pitch, add pauses, improve pronunciation, change speaking rate, adjust volume, and attribute multiple voices to a single document.

-To get started with text-to-speech, see the [quickstart](get-started-text-to-speech.md). Text-to-speech is available via the [Speech SDK](speech-sdk.md), the [REST API](rest-text-to-speech.md), and the [Speech CLI](spx-overview.md).

-> To convert text-to-speech with a no-code approach, try the [Audio Content Creation](how-to-audio-content-creation.md) tool in [Speech Studio](https://aka.ms/speechstudio/audiocontentcreation).

-Sample code for text-to-speech is available on GitHub. These samples cover text-to-speech conversion in most popular programming languages:

-* [Text-to-speech samples (SDK)](https://github.com/Azure-Samples/cognitive-services-speech-sdk)

-* [Text-to-speech samples (REST)](https://github.com/Azure-Samples/Cognitive-Speech-TTS)

-When you use the text-to-speech feature, you're billed for each character that's converted to speech, including punctuation. Although the SSML document itself is not billable, optional elements that are used to adjust how the text is converted to speech, like phonemes and pitch, are counted as billable characters. Here's a list of what's billable:

-* Text passed to the text-to-speech feature in the SSML body of the request

-* [REST API: Text-to-speech](rest-text-to-speech.md)

-## Connection closed or timeout

-There is a known issue on Windows 11 that might affect some types of Secure Sockets Layer (SSL) and Transport Layer Security (TLS) connections. These connections might have handshake failures. For developers, the affected connections are likely to send multiple frames followed by a partial frame with a size of less than 5 bytes within a single input buffer. If the connection fails, your app will receive the error such as, "USP error", "Connection closed", "ServiceTimeout", or "SEC_E_ILLEGAL_MESSAGE".

-There is an out of band update available for Windows 11 that fixes these issues. The update may be manually installed by following the instructions here:

-The issue started October 12th, 2022 and should be resolved via Windows update in November, 2022.

-1. The response text is turned into audio by the text-to-speech service, and streamed back to the client application for playback.

-Ensure that you use a [supported Azure region](regions.md#voice-assistants). The Direct Line Speech channel uses the text-to-speech service, which has neural and standard voices. Neural voices are used at [these Azure regions](regions.md#speech-service), and standard voices (retiring) are used at [these Azure regions](how-to-migrate-to-prebuilt-neural-voice.md).

-The bot that you've created will listen for and respond in English, with a default US English text-to-speech voice. However, you're not limited to using English or a default voice.

-You can choose from any of the languages mentioned in the [speech-to-text](language-support.md?tabs=stt) table. The following example changes the language to German.

-1. Open the Windows Voice Assistant Client app, select the **Settings** button (upper-right gear icon), and enter **de-de** in the **Language** field. This is the locale value mentioned in the [speech-to-text](language-support.md?tabs=stt) table.

-You can select the text-to-speech voice and control pronunciation if the bot specifies the reply in the form of a [Speech Synthesis Markup Language](speech-synthesis-markup.md) (SSML) instead of simple text. The echo bot doesn't use SSML, but you can easily modify the code to do that.

-* [Deploy to an Azure region that supports high-quality neural text-to-speech voices](./regions.md#speech-service).

-|[Speech-to-text](speech-to-text.md) | Voice assistants convert real-time audio into recognized text by using [speech-to-text](speech-to-text.md) from the Speech service. This text is available, as it's transcribed, to both your assistant implementation and your client application.

-|[Text-to-speech](text-to-speech.md) | Textual responses from your assistant are synthesized through [text-to-speech](text-to-speech.md) from the Speech service. This synthesis is then made available to your client application as an audio stream. Microsoft offers the ability to build your own custom, high-quality Neural Text to Speech (Neural TTS) voice that gives a voice to your brand.