Note
Please see Azure Cognitive Services for Speech documentation for the latest supported speech solutions.
What’s New in the Microsoft Speech Platform SDK 11
The following functionality has been recently added or updated in the Microsoft Speech Platform SDK.
VoiceXML Support
The managed-code Microsoft.Speech.VoiceXML namespace contains classes with which you can add the ability to process VoiceXML documents to your speech applications. The added functionality includes a VoiceXML runtime that interprets VoiceXML markup and a Browser class that allows users to communicate with a VoiceXML application. For more information, see VoiceXML Support (Microsoft.Speech).
Grammar Development Tools
The Microsoft Grammar Development Tools are a collection of command-line applications that you can use to test and optimize your voice applications without first deploying them to a live service. Using the Grammar Development Tools, you can do the following:
Create configuration files that select a speech recognition engine and set its properties.
Analyze grammars statically for errors.
Generate lists of phrases that a grammar will produce.
Find out whether a grammar contains a specific phrase.
Identify phrases in a grammar that may be confused with other phrases.
Perform simulated speech recognition on grammars.
Analyze recognition results for batches of utterances.
Compile grammar files to binary formats.
For more information, see Microsoft Grammar Development Tools.
Expanded Language Support
The Microsoft Speech Platform Runtime 11 now supports 26 languages. You must download a Runtime Language for each language that want to program against in your applications. There are separate Runtime Languages for speech recognition and speech synthesis. The version of the Runtime Languages that you use for development, for example version 11.0, must match the version of the Speech Platform Runtime that you installed.
Use the following link to download Runtime Languages for version 11 of the Speech Platform Runtime.