Merge branch 'MicrosoftDocs:main' into heidist-free

Author: HeidiSteen

articles/ai-services/.openpublishing.redirection.ai-services.json

@@ -400,6 +400,11 @@
       "redirect_url": "/azure/ai-services/speech-service/quickstarts/setup-platform",
       "redirect_document_id": false
     },
+    {
+      "source_path_from_root": "/articles/ai-services/speech-service/devices-sdk-release-notes.md",
+      "redirect_url": "/azure/ai-services/speech-service/release-notes",
+      "redirect_document_id": false
+    },
     {
       "source_path_from_root": "/articles/ai-services/anomaly-detector/how-to/postman.md",
       "redirect_url": "/azure/ai-services/anomaly-detector/overview",

articles/ai-services/custom-vision-service/overview.md

@@ -18,8 +18,7 @@ keywords: image recognition, image identifier, image recognition app, custom vis
 
 Azure AI Custom Vision is an image recognition service that lets you build, deploy, and improve your own **image identifier** models. An image identifier applies labels to images, according to their visual characteristics. Each label represents a classification or object. Custom Vision allows you to specify your own labels and train custom models to detect them.
 
-> [!TIP]
-> The Azure AI vision Image Analysis API, based on Florence foundational model, now supports custom models with few-shot learning capability. [Use Image Analysis 4.0](../computer-vision/how-to/model-customization.md) to create custom image identifier models using the latest technology from Azure. To migrate a Custom Vision project to the new Image Analysis 4.0 system, see the [Migration guide](../computer-vision/how-to/migrate-from-custom-vision.md). To compare the two services, see the [Comparison page](./concepts/compare-alternatives.md).
+
 
 
 You can use Custom Vision through a client library SDK, REST API, or through the [Custom Vision web portal](https://customvision.ai/). Follow a quickstart to get started.

articles/ai-services/language-service/conversational-language-understanding/concepts/best-practices.md

@@ -47,7 +47,7 @@ You also want to avoid mixing different schema designs. Don't build half of your
 
 ## Use standard training before advanced training
 
-[Standard training](../how-to/train-model.md#training-modes) is free and faster than advanced training. It can help you quickly understand the effect of changing your training set or schema while you build the model. After you're satisfied with the schema, consider using advanced training to get the best AIQ out of your model.
+[Standard training](../how-to/train-model.md#training-modes) is free and faster than advanced training. It can help you quickly understand the effect of changing your training set or schema while you build the model. After you're satisfied with the schema, consider using advanced training to get the best model quality.
 
 ## Use the evaluation feature
 
@@ -113,7 +113,7 @@ If you enable this feature, the utterance count of your training set increases.
 
 ## Address model overconfidence
 
-Customers can use the LoraNorm recipe version if the model is being incorrectly overconfident. An example of this behavior can be like the following scenario where the model predicts the incorrect intent with 100% confidence. This score makes the confidence threshold project setting unusable.
+Customers can use the LoraNorm traning configuration version if the model is being incorrectly overconfident. An example of this behavior can be like the following scenario where the model predicts the incorrect intent with 100% confidence. This score makes the confidence threshold project setting unusable.
 
 | Text |	Predicted intent |	Confidence score |
 |----|----|----|
@@ -243,7 +243,7 @@ curl --request POST \
 
 ## Address out-of-domain utterances
 
-Customers can use the newly updated recipe version `2024-08-01-preview` (previously `2024-06-01-preview`) if the model has poor AIQ on out-of-domain utterances. An example of this scenario with the default recipe can be like the following example where the model has three intents: `Sports`, `QueryWeather`, and `Alarm`. The test utterances are out-of-domain utterances and the model classifies them as `InDomain` with a relatively high confidence score.
+Customers can use the newly updated training configuration version `2024-08-01-preview` (previously `2024-06-01-preview`) if the model has poor quality on out-of-domain utterances. An example of this scenario with the default training configuration can be like the following example where the model has three intents: `Sports`, `QueryWeather`, and `Alarm`. The test utterances are out-of-domain utterances and the model classifies them as `InDomain` with a relatively high confidence score.
 
 | Text |	Predicted intent |	Confidence score |
 |----|----|----|
@@ -273,6 +273,6 @@ After the request is sent, you can track the progress of the training job in Lan
 
 Caveats:
 
-- The None score threshold for the app (confidence threshold below which `topIntent` is marked as `None`) when you use this recipe should be set to 0. This setting is used because this new recipe attributes a certain portion of the in-domain probabilities to out of domain so that the model isn't incorrectly overconfident about in-domain utterances. As a result, users might see slightly reduced confidence scores for in-domain utterances as compared to the prod recipe.
-- We don't recommend this recipe for apps with only two intents, such as `IntentA` and `None`, for example.
-- We don't recommend this recipe for apps with a low number of utterances per intent. We highly recommend a minimum of 25 utterances per intent.
+- The None score threshold for the app (confidence threshold below which `topIntent` is marked as `None`) when you use this training configuration should be set to 0. This setting is used because this new training configuration attributes a certain portion of the in-domain probabilities to out of domain so that the model isn't incorrectly overconfident about in-domain utterances. As a result, users might see slightly reduced confidence scores for in-domain utterances as compared to the prod training configuration.
+- We don't recommend this training configuration for apps with only two intents, such as `IntentA` and `None`, for example.
+- We don't recommend this training configuration for apps with a low number of utterances per intent. We highly recommend a minimum of 25 utterances per intent.

articles/ai-services/language-service/summarization/how-to/document-summarization.md

@@ -122,22 +122,11 @@ curl -i -X POST https://<your-language-resource-endpoint>/language/analyze-text/
     {
       "kind": "AbstractiveSummarization",
       "taskName": "Text Abstractive Summarization Task 1",
-      "parameters": {
-        "summaryLength": "short"
-      }
     }
   ]
 }
 '
 ```
-If you don't specify `summaryLength`, the model determines the summary length.
-
-### Using the summaryLength parameter
-For the `summaryLength` parameter, three values are accepted:
-* oneSentence: Generates a summary of mostly 1 sentence, with around 80 tokens.
-* short: Generates a summary of mostly 2-3 sentences, with around 120 tokens.
-* medium: Generates a summary of mostly 4-6 sentences, with around 170 tokens.
-* long: Generates a summary of mostly over 7 sentences, with around 210 tokens.
 
 2. Make the following changes in the command where needed:
     - Replace the value `your-language-resource-key` with your key.
@@ -222,7 +211,14 @@ The following cURL commands are executed from a BASH shell. Edit these commands
 
 The query-based text summarization API is an extension to the existing text summarization API.
 
-The biggest difference is a new `query` field in the request body (under `tasks` > `parameters` > `query`). Additionally, there's a new way to specify the preferred `summaryLength` in "buckets" of short/medium/long, which we recommend using instead of `sentenceCount`, especially when using abstractive. Below is an example request:
+The biggest difference is a new `query` field in the request body (under `tasks` > `parameters` > `query`).
+
+> [!TIP]
+> Query based summarization has some differentiation in the utilization of length control based on the type of query based summarization you're using:
+> - Query based extractive summarization supports length control by specifying sentenceCount.
+> - Query based abstractive summarization doesn't support length control.
+
+Below is an example request:
 
 ```bash
 curl -i -X POST https://<your-language-resource-endpoint>/language/analyze-text/jobs?api-version=2023-11-15-preview \
@@ -253,22 +249,93 @@ curl -i -X POST https://<your-language-resource-endpoint>/language/analyze-text/
       "kind": "ExtractiveSummarization",
       "taskName": "Query_based Extractive Summarization",
       "parameters": {
-          "query": "XYZ-code",
-          "sentenceCount": 3
+          "query": "XYZ-code"
       }
     }
   ]
 }
 '
 ```
 
-### Using the summaryLength parameter
+### Summary length control
+
+#### Using the summaryLength parameter in abstractive summarization
+
+If you don't specify `summaryLength`, the model determines the summary length.
+
 For the `summaryLength` parameter, three values are accepted:
 * oneSentence: Generates a summary of mostly 1 sentence, with around 80 tokens.
 * short: Generates a summary of mostly 2-3 sentences, with around 120 tokens.
 * medium: Generates a summary of mostly 4-6 sentences, with around 170 tokens.
 * long: Generates a summary of mostly over 7 sentences, with around 210 tokens.
 
+Below is an example request:
+
+```bash
+curl -i -X POST https://<your-language-resource-endpoint>/language/analyze-text/jobs?api-version=2023-04-01 \
+-H "Content-Type: application/json" \
+-H "Ocp-Apim-Subscription-Key: <your-language-resource-key>" \
+-d \
+' 
+{
+  "displayName": "Text Abstractive Summarization Task Example",
+  "analysisInput": {
+    "documents": [
+      {
+        "id": "1",
+        "language": "en",
+        "text": "At Microsoft, we have been on a quest to advance AI beyond existing techniques, by taking a more holistic, human-centric approach to learning and understanding. As Chief Technology Officer of Azure AI services, I have been working with a team of amazing scientists and engineers to turn this quest into a reality. In my role, I enjoy a unique perspective in viewing the relationship among three attributes of human cognition: monolingual text (X), audio or visual sensory signals, (Y) and multilingual (Z). At the intersection of all three, there’s magic—what we call XYZ-code as illustrated in Figure 1—a joint representation to create more powerful AI that can speak, hear, see, and understand humans better. We believe XYZ-code enables us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages. The goal is to have pretrained models that can jointly learn representations to support a broad range of downstream AI tasks, much in the way humans do today. Over the past five years, we have achieved human performance on benchmarks in conversational speech recognition, machine translation, conversational question answering, machine reading comprehension, and image captioning. These five breakthroughs provided us with strong signals toward our more ambitious aspiration to produce a leap in AI capabilities, achieving multi-sensory and multilingual learning that is closer in line with how humans learn and understand. I believe the joint XYZ-code is a foundational component of this aspiration, if grounded with external knowledge sources in the downstream AI tasks."
+      }
+    ]
+  },
+  "tasks": [
+    {
+      "kind": "AbstractiveSummarization",
+      "taskName": "Length controlled Abstractive Summarization",
+          "parameters": {
+          "sentenceLength": "short"
+      }
+    }
+  ]
+}
+'
+```
+
+#### Using the sentenceCount parameter in extractive summarization
+For the `sentenceCount` parameter, you can input a value 1-20 to indicate the desired number of output sentences.
+
+Below is an example request:
+
+```bash
+curl -i -X POST https://<your-language-resource-endpoint>/language/analyze-text/jobs?api-version=2023-11-15-preview \
+-H "Content-Type: application/json" \
+-H "Ocp-Apim-Subscription-Key: <your-language-resource-key>" \
+-d \
+' 
+{
+  "displayName": "Text Extractive Summarization Task Example",
+  "analysisInput": {
+    "documents": [
+      {
+        "id": "1",
+        "language": "en",
+        "text": "At Microsoft, we have been on a quest to advance AI beyond existing techniques, by taking a more holistic, human-centric approach to learning and understanding. As Chief Technology Officer of Azure AI services, I have been working with a team of amazing scientists and engineers to turn this quest into a reality. In my role, I enjoy a unique perspective in viewing the relationship among three attributes of human cognition: monolingual text (X), audio or visual sensory signals, (Y) and multilingual (Z). At the intersection of all three, there’s magic—what we call XYZ-code as illustrated in Figure 1—a joint representation to create more powerful AI that can speak, hear, see, and understand humans better. We believe XYZ-code enables us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages. The goal is to have pretrained models that can jointly learn representations to support a broad range of downstream AI tasks, much in the way humans do today. Over the past five years, we have achieved human performance on benchmarks in conversational speech recognition, machine translation, conversational question answering, machine reading comprehension, and image captioning. These five breakthroughs provided us with strong signals toward our more ambitious aspiration to produce a leap in AI capabilities, achieving multi-sensory and multilingual learning that is closer in line with how humans learn and understand. I believe the joint XYZ-code is a foundational component of this aspiration, if grounded with external knowledge sources in the downstream AI tasks."
+      }
+    ]
+  },
+"tasks": [
+    {
+      "kind": "ExtractiveSummarization",
+      "taskName": "Length controlled Extractive Summarization",
+      "parameters": {
+          "sentenceCount": "5"
+      }
+    }
+  ]
+}
+'
+```
+
 ## Service and data limits
 
 [!INCLUDE [service limits article](../../includes/service-limits-link.md)]

articles/ai-services/language-service/summarization/region-support.md

@@ -25,11 +25,14 @@ Some summarization features are only available in limited regions. More regions
 |North Europe      |✅                           |✅                                        |❌            |
 |East US           |✅                           |✅                                        |✅             |
 |East US 2         |✅                           |✅                                        |❌            |
+|Central US        |✅                           |✅                                        |❌            |
+|South Central US  |✅                           |✅                                        |❌            |
 |West US           |✅                           |✅                                        |❌            |
-|USNat West        |✅                           |✅                                        |❌            |
+|West US 2         |✅                           |✅                                        |❌            |
 |USNat East        |✅                           |✅                                        |❌            |
-|USSec West        |✅                           |✅                                        |❌            |
+|USNat West        |✅                           |✅                                        |❌            |
 |USSec East        |✅                           |✅                                        |❌            |
+|USSec West        |✅                           |✅                                        |❌            |
 |South UK          |✅                           |✅                                        |❌            |
 |Southeast Asia    |✅                           |✅                                        |❌            |
 |Australia East    |✅                           |✅                                        |❌            |

articles/ai-services/openai/concepts/model-retirements.md

@@ -4,7 +4,7 @@ titleSuffix: Azure OpenAI
 description: Learn about the model deprecations and retirements in Azure OpenAI.
 ms.service: azure-ai-openai
 ms.topic: conceptual
-ms.date: 09/09/2024
+ms.date: 09/12/2024
 ms.custom: 
 manager: nitinme
 author: mrbullwinkle
@@ -91,9 +91,9 @@ These models are currently available for use in Azure OpenAI Service.
 
 | Model | Version | Retirement date | Suggested replacements |
 | ---- | ---- | ---- | --- |
-| `gpt-35-turbo` | 0301 | January 27, 2025<br><br> Deployments set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 15, 2024.   | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini`  |
-| `gpt-35-turbo`<br>`gpt-35-turbo-16k` | 0613 | January 27, 2025 <br><br> Deployments set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 15, 2024.  | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini`|
-| `gpt-35-turbo` | 1106 | No earlier than Nov 17, 2024 <br><br> Deployments set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 15, 2024. | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini` |
+| `gpt-35-turbo` | 0301 | January 27, 2025<br><br> Deployments set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 13, 2024.   | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini`  |
+| `gpt-35-turbo`<br>`gpt-35-turbo-16k` | 0613 | January 27, 2025 <br><br> Deployments set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 13, 2024.  | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini`|
+| `gpt-35-turbo` | 1106 | No earlier than Nov 17, 2024 <br><br> Deployments set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 13, 2024. | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini` |
 | `gpt-35-turbo` | 0125 | No earlier than Feb 22, 2025 | `gpt-4o-mini` |
 | `gpt-4`<br>`gpt-4-32k` | 0314 | June 6, 2025 | `gpt-4o` |
 | `gpt-4`<br>`gpt-4-32k` | 0613 | June 6, 2025 | `gpt-4o` |
@@ -115,7 +115,7 @@ These models are currently available for use in Azure OpenAI Service.
 
 | Model | Current default version | New default version | Default upgrade date |
 |---|---|---|---|
-| `gpt-35-turbo` | 0301 | 0125 | Deployments of versions `0301`, `0613`, and `1106` set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 15, 2024.|
+| `gpt-35-turbo` | 0301 | 0125 | Deployments of versions `0301`, `0613`, and `1106` set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 13, 2024.|
 
 
 
@@ -158,6 +158,10 @@ If you're an existing customer looking for information about these models, see [
 
 ## Retirement and deprecation history
 
+## September 12, 2024
+
+* `gpt-35-turbo` (0301), (0613), (1106) and `gpt-35-turbo-16k` (0613) auto-update to default upgrade date updated to November 13, 2024.
+
 ## September 9, 2024
 
 * `gpt-35-turbo` (0301) and (0613) retirement changed to January 27, 2025.