Larry Magid did an article on speech recognition last weekend on CBS News. In the article he talked about his experience with both Dragon Naturally Speaking (DNS) version 9, as well as Windows Speech Recognition (WSR) included in the upcoming Windows Vista Operating System.

He had several positive things to say about Windows Speech Recognition as compared with Dragon's product:

  • “I found both the Vista application and NaturallySpeaking to be accurate and easy to use”
  • “To their credit both programs are accurate and intelligent in that they recognize not only words but also allow you to issue commands.”
  • “After using both for a few days, each got to the point where they recognized about 99% of my words”
  • “Having been temporarily disabled, I can easily see how this could be a wonderful tool for anyone with a permanent or long-term disability as well as those suffering from repetitive strain injuries or other hand problems.”

It's nice to hear that he thinks Windows Speech Recognition is in most ways just as good as Dragon Naturally Speaking. Especially on the accuracy part. DNS v9 got a lot of credit recently for their better accuracy. It's nice to know that Larry sees the two products as equal in that regard.

Larry did say DNS was a more sophisticated product, partly due to their multiple language support, additional vocabularies, and support for handheld digital recorders.

However, WSR does actually support multiple languages. It's a common misconception that we don't. In Windows Vista we support English, UK English, German, French, Spanish, Japanese, Simplified Chinese, and Traditional Chinese.

WSR also has support for additional vocabularies; neither we nor our partners have built and shipped any yet, though. I hope to see some announcements here in the future. :-)

It's true that we don't have built-in support for digital recorders, but we do have something even better, IMO. A public API that's freely distributed with the Windows SDK: SAPI and System.Speech.

Developers can use these APIs to build world class applications that integrate speech in new and innovative ways in their applications. All it would take is a motivated third party to build and deploy an application that worked with the digital recorders that are in the market today.