INFO: SAPI 5.1 Speech Recognition Engine Language Support (318741)

SUMMARY

The typical functions of speech recognition programs include the following tasks:

Capture sound input.
Convert sound input to basic language units or phonemes.
Construct words from phonemes.
Contextually analyze the words to make sure that words that sound alike are spelled correctly in their context.

Speech recognition (SR) engines are the software drivers that perform the following tasks:

Convert the acoustic signal into text.
Contain the recognized speech.
Pass the recognized speech to your application.

Microsoft Speech Recognition Engines

Microsoft SR engines support continuous speech, which means that you can speak naturally into a microphone and not pause after every word that you speak.

Modes of Speech Recognition

Microsoft SR engines currently support two modes of speech recognition:

Dictation. To enter text, such as an e-mail message, the user reads directly to the microphone.
Command and Control. To initiate actions, such as Send Mail or Italicize That, the user speaks commands or asks questions.

Supported Languages

Microsoft SR engines currently support these languages:

Simplified Chinese
English
Japanese

MORE INFORMATION

Microsoft has no immediate plan to provide support for other languages. However, Microsoft is working closely with third-party speech engine vendors. For more information and links to third-party SR providers, visit the following Microsoft Web site:

.NET Speech - Third Party Products http://www.microsoft.com/speech/evaluation/thirdparty/engines.asp

Microsoft is investigating the feasibility of speech recognition on small devices that are based on Microsoft Windows CE, but does not currently offer an SR engine that will run on Windows CE.