The document discusses different ways of communicating with computers, including traditional methods like keyboards and mice as well as modern methods like touch, speech, and motion. It focuses on speech synthesis and recognition, how they work, and their applications. Speech synthesis converts text to audio while recognition does the opposite. Examples of speech technology uses include movies, translation, language learning, mobiles, robotics, and games. The document also covers the Windows Speech API and its history, changes in newer versions of Windows to improve speech recognition, and a demo of speech synthesis and recognition in .NET.
21. History of SAPI
1994: SAPI 1.0
Windows 95 / Windows NT
1998: SAPI 4.0
C++ wrapper classes
ActiveX for Visual basic
22. History of SAPI
1994: SAPI 1.0
Windows 95 / Windows NT
1998: SAPI 4.0
C++ wrapper classes
ActiveX for Visual basic
2006: SAPI 5.3
Windows Vista
23. Ιστορία τοσ SAPI
1994: SAPI 1.0
Windows 95 / Windows NT
1998: SAPI 4.0
C++ wrapper classes
ActiveX for Visual basic
2006: SAPI 5.3
Windows Vista
2009: SAPI 5.4
Windows 7
25. Changes in Windows Vista & 7
Upgraded Speech Recognition engine
Separate application with its own GUI
26. Changes in Windows Vista & 7
Upgraded Speech Recognition engine
Separate application with its own GUI
Checks the UI operation
27. Changes in Windows Vista & 7
Upgraded Speech Recognition engine
Separate application with its own GUI
Checks the UI operation
Supports more languages -
English US & UK, Chinese traditional & simplified,
Japanese, German, French, Spanish
28. Changes in Windows Vista & 7
Upgraded Speech Recognition engine
Separate application with its own GUI
Checks the UI operation
Supports more languages -
English US & UK, Chinese traditional & simplified,
Japanese, German, French, Spanish
Managed code speech API (.ΝΕΤ 3.0)
29. What we use
Technologies
• .NET Framework 4.0
• C# programming language
• Windows Presentation Foundation
Tools
• Windows 7
• Visual Studio 2010
• FREE @ MSDNAA
30. Windows Speech Synthesis
Converts words into voice
Internet settings like:
intensity
Pronunciation (voice)
Introducing WAV files
By default, uses Microsoft Anna
32. Windows Speech Recognition
Uses machine learning algorithms
Continuously Trained
Trains using the user’s voice
Can be used for remote control of the
PC