[Accessibility] Speech Recognition Article

[Jared Wright]

Aside from being generally   interesting research, I thought the 
approach of  using Youtube videos and the countless other sources of 
spoken word in our mediacentric internet as another source of real-world 
speech samples to train intelligent speech recognition  algorithms was 
particularly  intriguing. With some of the traffic on this list in 
recent weeks, I thought I'd pass it along.

http://discovere.binghamton.edu/faculty-spotlights/zahorian-3249.html