Merge pull request #107978 from IEvangelist/migration

Fixes for prog lang support of SDK in migration doc

Author: Jak-MS

articles/cognitive-services/Speech-Service/how-to-migrate-from-bing-speech.md

@@ -9,7 +9,7 @@ manager: nitinme
 ms.service: cognitive-services
 ms.subservice: speech-service
 ms.topic: conceptual
-ms.date: 01/21/2020
+ms.date: 03/17/2020
 ms.author: nitinme
 
 # Customer intent: As a developer currently using the deprecated Bing Speech, I want to learn the differences between Bing Speech and the Speech service, so that I can migrate my application to the Speech service.
@@ -35,25 +35,25 @@ The [Speech SDK](speech-sdk.md) is a functional replacement for the Bing Speech
 
 The Speech service is largely similar to Bing Speech, with the following differences.
 
-Feature | Bing Speech | Speech service | Details
--|-|-|-
-C++ SDK | :heavy_minus_sign: | :heavy_check_mark: | Speech service supports Windows and Linux.
-Java SDK | :heavy_check_mark: | :heavy_check_mark: | Speech service supports Android and Speech Devices.
-C# SDK | :heavy_check_mark: | :heavy_check_mark: | Speech service supports Windows 10, Universal Windows Platform (UWP), and .NET Standard 2.0.
-Continuous speech recognition | 10 minutes | Unlimited (with SDK) | Both Bing Speech and Speech service WebSockets protocols support up to 10 minutes per call. However, the Speech SDK automatically reconnects on timeout or disconnect.
-Partial or interim results | :heavy_check_mark: | :heavy_check_mark: | With WebSockets protocol or SDK.
-Custom speech models | :heavy_check_mark: | :heavy_check_mark: | Bing Speech requires a separate Custom Speech subscription.
-Custom voice fonts | :heavy_check_mark: | :heavy_check_mark: | Bing Speech requires a separate Custom Voice subscription.
-24-KHz voices | :heavy_minus_sign: | :heavy_check_mark:
-Speech intent recognition | Requires separate LUIS API call | Integrated (with SDK) |  You can use a LUIS key with the Speech service.
-Simple intent recognition | :heavy_minus_sign: | :heavy_check_mark:
-Batch transcription of long audio files | :heavy_minus_sign: | :heavy_check_mark:
-Recognition mode | Manual via endpoint URI | Automatic | Recognition mode is not available in the Speech service.
-Endpoint locality | Global | Regional | Regional endpoints improve latency.
-REST APIs | :heavy_check_mark: | :heavy_check_mark: | The Speech service REST APIs are compatible with Bing Speech (different endpoint). REST APIs support text-to-speech and limited speech-to-text functionality.
-WebSockets protocols | :heavy_check_mark: | :heavy_check_mark: | The Speech service WebSockets API is compatible with Bing Speech (different endpoint). Migrate to the Speech SDK if possible, to simplify your code.
-Service-to-service API calls | :heavy_check_mark: | :heavy_minus_sign: | Provided in Bing Speech via the C# Service Library.
-Open source SDK | :heavy_check_mark: | :heavy_minus_sign: |
+| Feature | Bing Speech | Speech service | Details |
+|--|--|--|--|
+| C# SDK | :heavy_check_mark: | :heavy_check_mark: | Speech service supports Windows 10, Universal Windows Platform (UWP), and .NET Standard 2.0. |
+| C++ SDK | :heavy_minus_sign: | :heavy_check_mark: | Speech service supports Windows and Linux. |
+| Java SDK | :heavy_check_mark: | :heavy_check_mark: | Speech service supports Android and Speech Devices. |
+| Continuous speech recognition | 10 minutes | Unlimited (with SDK) | Both Bing Speech and Speech service WebSockets protocols support up to 10 minutes per call. However, the Speech SDK automatically reconnects on timeout or disconnect. |
+| Partial or interim results | :heavy_check_mark: | :heavy_check_mark: | With WebSockets protocol or SDK. |
+| Custom speech models | :heavy_check_mark: | :heavy_check_mark: | Bing Speech requires a separate Custom Speech subscription. |
+| Custom voice fonts | :heavy_check_mark: | :heavy_check_mark: | Bing Speech requires a separate Custom Voice subscription. |
+| 24-KHz voices | :heavy_minus_sign: | :heavy_check_mark: |
+| Speech intent recognition | Requires separate LUIS API call | Integrated (with SDK) | You can use a LUIS key with the Speech service. |
+| Simple intent recognition | :heavy_minus_sign: | :heavy_check_mark: |
+| Batch transcription of long audio files | :heavy_minus_sign: | :heavy_check_mark: |
+| Recognition mode | Manual via endpoint URI | Automatic | Recognition mode is not available in the Speech service. |
+| Endpoint locality | Global | Regional | Regional endpoints improve latency. |
+| REST APIs | :heavy_check_mark: | :heavy_check_mark: | The Speech service REST APIs are compatible with Bing Speech (different endpoint). REST APIs support text-to-speech and limited speech-to-text functionality. |
+| WebSockets protocols | :heavy_check_mark: | :heavy_check_mark: | The Speech service WebSockets API is compatible with Bing Speech (different endpoint). Migrate to the Speech SDK if possible, to simplify your code. |
+| Service-to-service API calls | :heavy_check_mark: | :heavy_minus_sign: | Provided in Bing Speech via the C# Service Library. |
+| Open-source SDK | :heavy_check_mark: | :heavy_minus_sign: |
 
 The Speech service uses a time-based pricing model (rather than a transaction-based model). See [Speech service pricing](https://azure.microsoft.com/pricing/details/cognitive-services/speech-services/) for details.
 
@@ -65,9 +65,7 @@ The Speech service [REST APIs](rest-apis.md) are compatible with the Bing Speech
 
 The Speech service WebSockets protocols are also compatible with those used by Bing Speech. We recommend that for new development, you use the Speech SDK rather than WebSockets. It's a good idea to migrate existing code to the SDK as well. However, as with the REST APIs, existing code that uses Bing Speech via WebSockets requires only a change in endpoint and an updated key.
 
-If you're using a Bing Speech client library for a specific programming language, migrating to the [Speech SDK](speech-sdk.md) requires changes to your application, because the API is different. The Speech SDK can make your code simpler, while also giving you access to new features.
-
-Currently, the Speech SDK supports C# ([details here](https://aka.ms/csspeech)), Java (Android and custom devices), Objective C (iOS), C++ (Windows and Linux), and JavaScript. APIs on all platforms are similar, easing multi-platform development.
+If you're using a Bing Speech client library for a specific programming language, migrating to the [Speech SDK](speech-sdk.md) requires changes to your application, because the API is different. The Speech SDK can make your code simpler, while also giving you access to new features. The Speech SDK is available in a wide variety of programming languages. APIs on all platforms are similar, easing multi-platform development.
 
 The Speech service doesn't offer a global endpoint. Determine if your application functions efficiently when it uses a single regional endpoint for all of its traffic. If not, use geolocation to determine the most efficient endpoint. You need a separate Speech service subscription in each region you use.