Speech APIs and SDK
Create your own applications that can use speech input and output. With just a couple of lines of code you can make your Microsoft Windows® applications listen and speak.
In addition to developing pioneering spoken language technology, Microsoft provides application programming interfaces (API’s) that allow Windows applications to use speech input and voice output. These are the same programming interfaces used by Microsoft desktop and server applications such as Windows® Speech Recognition, Microsoft® Narrator, Microsoft® Speech Server, and the spoken driving instructions included in Microsoft® Streets and Trips and Microsoft® MapPoint.
The API provides a method of communication and coordination between the application and one or more speech engines. Microsoft makes available or includes a high-quality speech recognition engine and text-to-speech (TTS) engine for all recent Windows versions.
For native applications, the Windows platform includes the Speech API (SAPI). This COM-based API includes dozens of objects and interfaces that can be used by applications to listen for speech, recognize content, process spoken commands, and speak text. SAPI is most easily used from applications written using C, C++, Visual Basic or other programming languages. It works with development frameworks such as the Microsoft Foundation Classes (MFC) and the Active Template Library (ATL). The Speech API is included in all Microsoft Windows versions since Windows 98. Microsoft Windows XP and Windows Server 2003 include SAPI version 5.1. Microsoft Windows Vista and the forthcoming Windows Server 2008 includes SAPI version 5.3. For managed code written for the .NET platform, Microsoft includes the Managed Speech API in the .NET Framework 3.0. The Managed Speech API includes the System.Speech namespace which makes developing speech-enabled applications in C# and Visual Basic.NET incredibly easy and robust.
Speech Engines
Windows Vista includes in all editions our latest speech recognition engine, version 8.0. Available in US English , UK English, French, German, Spanish, Japanese, Traditional Chinese and Simplified Chinese. The engine is included with the corresponding Windows Vista language. Users can install multiple languages by purchasing Windows Multilingual User Interface Packs. For Windows XP, version 6.1 of our Speech Recognition engine is included in Windows XP Tablet Edition, and Microsoft Office 2003. Developers of speech input applications for Windows XP can also contact Microsoft regarding redistributing the version 6.1 engine with their own applications. Spoken output is accomplished using one of the built-in TTS engines in Microsoft Windows or a third-party engine that supports the SAPI 5.x interfaces. Windows XP included the Microsoft Sam voice and Windows Vista includes next-generation TTS technology in the Microsoft Anna (US English) and Microsoft Lili (Chinese) voices. Office 2002 and Office 2003 also included TTS engines. Many third-party engines in a wide range of languages are also available.
Speech SDK
Application software developers can start using the Speech APIs immediately by downloading the Windows SDK. The latest SDK contains updated documentation on SAPI 5.3 and the Managed Speech API in .NET Framework 3.0. You can read the documentation on-line from MSDN or download the complete SDK.
