Albert asked me this question:

Hi Richard, nice blog. I liked the 'Devil' video. I have to give a pretty technical explanation taking speech recognition from linguistics through the technical side of how it works and point to a working demo. Any ideas? It's for a language technologies and applications course.

If you're looking for general information, I would start on the MSR speech page. Some of that information is old (argh! wish we had more time to update it!) but fortunately the fundamental technology hasn't changed much, so for what you need it shouldn't matter.  Also be sure to get a copy of Spoken Language Processing, the best text book ever written about SR (hi boss!)

Vista, of course, is the best place to start if you want a working demo. If that's not possible, take a look at the Tellme Studio site, which will let you whip up a nice phone demo very quickly.

You probably came to my blog via an internet search, so now that you're here I encourage you to dig deeper through the archives for other pointers to presentations (maybe look at Sept 05) and other docs I've pointed to over the years, like the SpeechWiki or the MS-Speech group on Yahoo.  If you have questions about Windows Speech Recognition, then go to Rob's Rhapsody, which is best blog for that.