Merge pull request #225639 from goergenj/docs-editor/releasenotes-1675114658

Disconnected CSTT Container GA release docs update

Author: denrea

articles/cognitive-services/Speech-Service/includes/release-notes/release-notes-containers.md

@@ -6,7 +6,6 @@ ms.date: 11/29/2022
 ms.author: eur
 ---
 
-
 ### 2023-January release
 
 #### New container versions
@@ -20,6 +19,8 @@ Fix Hypothesis mode issue
 
 Fix HTTP Proxy issue
 
+Custom Speech-to-Text container disconnected mode
+
 Add CNV Disconnected container support to TTS Frontend
 
 Add support for these locale-voices:
@@ -134,3 +135,4 @@ Regular monthly updates including security upgrades and vulnerability fixes.
 Add support for these prebuilt neural voices: `am-et-amehaneural`, `am-et-mekdesneural`, `so-so-muuseneural` and `so-so-ubaxneural`.
 
 Regular monthly updates including security upgrades and vulnerability fixes.
+

articles/cognitive-services/Speech-Service/releasenotes.md

@@ -19,6 +19,7 @@ Azure Cognitive Service for Speech is updated on an ongoing basis. To stay up-to
 
 ## Recent highlights
 
+* Custom Speech-to-Text container disconnected mode was released in January 2023.
 * Speech SDK 1.25.0 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.

articles/cognitive-services/Speech-Service/speech-container-configuration.md

@@ -129,8 +129,9 @@ The following Docker examples are for the Speech container.
 
 ### Basic example for Speech-to-text
 
+
 ```Docker
-docker run --rm -it -p 5000:5000 --memory 4g --cpus 4 \
+docker run --rm -it -p 5000:5000 --memory 8g --cpus 4 \
 mcr.microsoft.com/azure-cognitive-services/speechservices/speech-to-text \
 Eula=accept \
 Billing={ENDPOINT_URI} \
@@ -139,8 +140,9 @@ ApiKey={API_KEY}
 
 ### Logging example for Speech-to-text
 
+
 ```Docker
-docker run --rm -it -p 5000:5000 --memory 4g --cpus 4 \
+docker run --rm -it -p 5000:5000 --memory 8g --cpus 4 \
 mcr.microsoft.com/azure-cognitive-services/speechservices/custom-speech-to-text \
 Eula=accept \
 Billing={ENDPOINT_URI} \
@@ -152,8 +154,9 @@ Logging:Console:LogLevel:Default=Information
 
 ### Basic example for Custom Speech-to-text
 
+
 ```Docker
-docker run --rm -it -p 5000:5000 --memory 4g --cpus 4 \
+docker run --rm -it -p 5000:5000 --memory 8g --cpus 4 \
 -v {VOLUME_MOUNT}:/usr/local/models \
 mcr.microsoft.com/azure-cognitive-services/speechservices/custom-speech-to-text \
 ModelId={MODEL_ID} \
@@ -164,8 +167,9 @@ ApiKey={API_KEY}
 
 ### Logging example for Custom Speech-to-text
 
+
 ```Docker
-docker run --rm -it -p 5000:5000 --memory 4g --cpus 4 \
+docker run --rm -it -p 5000:5000 --memory 8g --cpus 4 \
 -v {VOLUME_MOUNT}:/usr/local/models \
 mcr.microsoft.com/azure-cognitive-services/speechservices/custom-speech-to-text \
 ModelId={MODEL_ID} \
@@ -202,8 +206,9 @@ Logging:Console:LogLevel:Default=Information
 
 ### Basic example for Speech language identification
 
+
 ```Docker
-docker run --rm -it -p 5000:5000 --memory 12g --cpus 6 \
+docker run --rm -it -p 5000:5000 --memory 1g --cpus 1 \
 mcr.microsoft.com/azure-cognitive-services/speechservices/language-detection \
 Eula=accept \
 Billing={ENDPOINT_URI} \
@@ -212,8 +217,9 @@ ApiKey={API_KEY}
 
 ### Logging example for Speech language identification
 
+
 ```Docker
-docker run --rm -it -p 5000:5000 --memory 12g --cpus 6 \
+docker run --rm -it -p 5000:5000 --memory 1g --cpus 1 \
 mcr.microsoft.com/azure-cognitive-services/speechservices/language-detection \
 Eula=accept \
 Billing={ENDPOINT_URI} \
@@ -226,3 +232,4 @@ Logging:Console:LogLevel:Default=Information
 
 - Review [How to install and run containers](speech-container-howto.md)
 
+

articles/cognitive-services/Speech-Service/speech-container-howto.md

@@ -53,19 +53,21 @@ You must meet the following prerequisites before you use Speech service containe
 
 The following table describes the minimum and recommended allocation of resources for each Speech container:
 
-| Container | Minimum | Recommended |
-|-----------|---------|-------------|
-| Speech-to-text | 4 core, 4-GB memory | 8 core, 6-GB memory |
-| Custom speech-to-text | 4 core, 4-GB memory | 8 core, 6-GB memory |
-| Speech language identification | 1 core, 1-GB memory | 1 core, 1-GB memory |
-| Neural text-to-speech | 6 core, 12-GB memory | 8 core, 16-GB memory |
+| Container | Minimum | Recommended |Speech Model|
+|-----------|---------|-------------| -------- |
+| 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|
+| Speech language identification | 1 core, 1-GB memory | 1 core, 1-GB memory |n/a|
+| Neural text-to-speech | 6 core, 12-GB memory | 8 core, 16-GB memory |n/a|
 
 Each core must be at least 2.6 gigahertz (GHz) or faster.
 
 Core and memory correspond to the `--cpus` and `--memory` settings, which are used as part of the `docker run` command.
 
 > [!NOTE]
-> The minimum and recommended allocations are based on Docker limits, *not* the host machine resources. For example, speech-to-text containers memory map portions of a large language model. We recommend that the entire file should fit in memory, which is an additional 4 to 6 GB. Also, the first run of either container might take longer because models are being paged into memory.
+> The minimum and recommended allocations are based on Docker limits, *not* the host machine resources.
+> 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 modesl (see above table).
+> Also, the first run of either container might take longer because models are being paged into memory.
 
 ### Advanced Vector Extension support
 
@@ -127,7 +129,6 @@ To use the latest version of the container, you can use the `latest` tag. You ca
 | Speech language identification | `mcr.microsoft.com/azure-cognitive-services/speechservices/language-detection:latest` |
 
 ***
-
 [!INCLUDE [Tip for using docker list](../../../includes/cognitive-services-containers-docker-list-tip.md)]
 
 ### Get the container image with docker pull
@@ -217,7 +218,6 @@ docker pull mcr.microsoft.com/azure-cognitive-services/speechservices/language-d
 ```
 
 ***
-
 ## Use the container
 
 After the container is on the [host computer](#host-computer-requirements-and-recommendations), use the following process to work with the container.
@@ -237,7 +237,7 @@ Use the [docker run](https://docs.docker.com/engine/reference/commandline/run/)
 To run the standard speech-to-text container, execute the following `docker run` command:
 
 ```bash
-docker run --rm -it -p 5000:5000 --memory 4g --cpus 4 \
+docker run --rm -it -p 5000:5000 --memory 8g --cpus 4 \
 mcr.microsoft.com/azure-cognitive-services/speechservices/speech-to-text \
 Eula=accept \
 Billing={ENDPOINT_URI} \
@@ -247,7 +247,7 @@ ApiKey={API_KEY}
 This command:
 
 * Runs a *speech-to-text* container from the container image.
-* Allocates 4 CPU cores and 4 GB of memory.
+* Allocates 4 CPU cores and 8 GB of memory.
 * Exposes TCP port 5000 and allocates a pseudo-TTY for the container.
 * Automatically removes the container after it exits. The container image is still available on the host computer.
 
@@ -364,7 +364,7 @@ The following table represents the various `docker run` parameters and their cor
 To run the custom speech-to-text container, execute the following `docker run` command:
 
 ```bash
-docker run --rm -it -p 5000:5000 --memory 4g --cpus 4 \
+docker run --rm -it -p 5000:5000 --memory 8g --cpus 4 \
 -v {VOLUME_MOUNT}:/usr/local/models \
 mcr.microsoft.com/azure-cognitive-services/speechservices/custom-speech-to-text \
 ModelId={MODEL_ID} \
@@ -376,7 +376,7 @@ ApiKey={API_KEY}
 This command:
 
 * Runs a custom speech-to-text container from the container image.
-* Allocates 4 CPU cores and 4 GB of memory.
+* Allocates 4 CPU cores and 8 GB of memory.
 * Loads the custom speech-to-text model from the volume input mount, for example, *C:\CustomSpeech*.
 * Exposes TCP port 5000 and allocates a pseudo-TTY for the container.
 * Downloads the model given the `ModelId` (if not found on the volume mount).
@@ -513,14 +513,15 @@ docker run --rm -v ${HOME}:/root -ti antsu/on-prem-client:latest ./speech-to-tex
 Increasing the number of concurrent calls can affect reliability and latency. For language identification, we recommend a maximum of four concurrent calls using 1 CPU with 1 GB of memory. For hosts with 2 CPUs and 2 GB of memory, we recommend a maximum of six concurrent calls.
 
 ***
-
 > [!IMPORTANT]
 > The `Eula`, `Billing`, and `ApiKey` options must be specified to run the container. Otherwise, the container won't start. For more information, see [Billing](#billing).
 
 ## Run the container in disconnected environments
 
 You must request access to use containers disconnected from the internet. For more information, see [Request access to use containers in disconnected environments](../containers/disconnected-containers.md#request-access-to-use-containers-in-disconnected-environments).
 
+For Speech Service container configuration, see [Disconnected containers](../containers/disconnected-containers.md#speech-containers).
+
 ## Query the container's prediction endpoint
 
 > [!NOTE]
@@ -644,7 +645,6 @@ speech_config.set_service_property(
 ```
 
 ---
-
 If you want to completely disable sentiment analysis, add a `false` value to `sentimentanalysis.enabled`.
 
 ```python
@@ -709,3 +709,5 @@ In this article, you learned concepts and workflow for how to download, install,
 * Review [configure containers](speech-container-configuration.md) for configuration settings.
 * Learn how to [use Speech service containers with Kubernetes and Helm](speech-container-howto-on-premises.md).
 * Use more [Cognitive Services containers](../cognitive-services-container-support.md).
+
+

articles/cognitive-services/containers/disconnected-containers.md

@@ -161,9 +161,11 @@ If you're using the [Translator container](../translator/containers/translator-h
 -e TRANSLATORSYSTEMCONFIG=/path/to/model/config/translatorsystemconfig.json
 ```
 
-#### Speech-to-text, Custom Speech-to-Text and Neural text-to-speech containers
+#### Speech containers
 
-The [Speech-to-Text](../speech-service/speech-container-howto.md?tabs=stt), [Custom Speech-to-Text](../speech-service/speech-container-howto.md?tabs=cstt) and [Neural Text-to-Speech](../speech-service/speech-container-howto.md?tabs=ntts) containers provide a default directory for writing the license file and billing log at runtime. The default directories are /license and /output respectively. 
+# [Speech-to-text](#tab/stt)
+
+The [Speech-to-Text](../speech-service/speech-container-howto.md?tabs=stt) container provides a default directory for writing the license file and billing log at runtime. The default directories are /license and /output respectively. 
 
 When you're mounting these directories to the container with the `docker run -v` command, make sure the local machine directory is set ownership to `user:group nonroot:nonroot` before running the container.
 
@@ -173,6 +175,52 @@ Below is a sample command to set file/directory ownership.
 sudo chown -R nonroot:nonroot <YOUR_LOCAL_MACHINE_PATH_1> <YOUR_LOCAL_MACHINE_PATH_2> ...
 ```
 
+# [Neural Text-to-Speech](#tab/ntts)
+
+The [Neural Text-to-Speech](../speech-service/speech-container-howto.md?tabs=ntts) container provides a default directory for writing the license file and billing log at runtime. The default directories are /license and /output respectively. 
+
+When you're mounting these directories to the container with the `docker run -v` command, make sure the local machine directory is set ownership to `user:group nonroot:nonroot` before running the container.
+
+Below is a sample command to set file/directory ownership.
+
+```bash
+sudo chown -R nonroot:nonroot <YOUR_LOCAL_MACHINE_PATH_1> <YOUR_LOCAL_MACHINE_PATH_2> ...
+```
+
+# [Custom Speech-to-Text](#tab/cstt)
+
+In order to prepare and configure the Custom Speech-to-Text container you will need two separate speech resources:
+
+1. A regular Azure Speech Service resource which is either configured to use a "**S0 - Standard**" pricing tier or a "**Speech to Text (Custom)**" commitment tier pricing plan. This will be used to train, download, and configure your custom speech models for use in your container.
+1. An Azure Speech Service resource which is configured to use the "**DC0 Commitment (Disconnected)**" pricing plan. This is used to download your disconnected container license file required to run the container in disconnected mode.
+   
+To download all the required models into your Custom Speech-to-Text container follow the instructions for Custom Speech-to-Text containers on the [Install and run Speech containers](../speech-service/speech-container-howto.md?tabs=cstt) page and use the  speech resource in step 1.
+
+After all required models have been downloaded to your host computer, you need to download the disconnected license file using the instructions in the above chapter, titled [Configure the container to be run in a disconnected environment](./disconnected-containers.md#configure-the-container-to-be-run-in-a-disconnected-environment), using the Speech resource from step 2.
+
+To run the container in disconnected mode, follow the instructions from above chapter titled [Run the container in a disconnected environment](./disconnected-containers.md#run-the-container-in-a-disconnected-environment) and add an additional `-v` parameter to mount the directory containing your custom speech model.
+
+Example for running a Custom Speech-to-Text container in disconnected mode:
+```bash
+docker run --rm -it -p 5000:5000 --memory {MEMORY_SIZE} --cpus {NUMBER_CPUS} \ 
+-v {LICENSE_MOUNT} \ 
+-v {OUTPUT_PATH} \
+-v {MODEL_PATH} \
+{IMAGE} \
+eula=accept \
+Mounts:License={CONTAINER_LICENSE_DIRECTORY}
+Mounts:Output={CONTAINER_OUTPUT_DIRECTORY}
+```
+
+The [Custom Speech-to-Text](../speech-service/speech-container-howto.md?tabs=cstt) container provides a default directory for writing the license file and billing log at runtime. The default directories are /license and /output respectively. 
+
+When you're mounting these directories to the container with the `docker run -v` command, make sure the local machine directory is set ownership to `user:group nonroot:nonroot` before running the container.
+
+Below is a sample command to set file/directory ownership.
+
+```bash
+sudo chown -R nonroot:nonroot <YOUR_LOCAL_MACHINE_PATH_1> <YOUR_LOCAL_MACHINE_PATH_2> ...
+```
 ## Usage records
 
 When operating Docker containers in a disconnected environment, the container will write usage records to a volume where they're collected over time. You can also call a REST endpoint to generate a report about service usage.
@@ -255,3 +303,7 @@ If you run the container with an output mount and logging enabled, the container
 ## Next steps
 
 [Azure Cognitive Services containers overview](../cognitive-services-container-support.md)
+
+
+
+