[FoRK] Cheap Electronics Dissection Project
Kragen Javier Sitaker <
kragen at pobox.com
> on >
Fri Nov 17 10:21:03 PST 2006
On Thu, 26 Oct 2006 12:17:49 +0200, Dave Long wrote:
> > There's good voice-recognition software that supposedly needs about
> > 15MIPS (ViaVoice) to run in real-time, and you may not even need
> > real-time. I don't know how much bandwidth to how much memory
> > speech-recognition needs, though.
> Looks like a fair amount: ViaVoice not only specs system memory
> (64-96M) but also 256K of L2 cache. There are some people nearby
> working on automatic segmentation/speaker recognition/tagging and
> search, etc., and they use serious crunch for their applications, but
> maybe they'd have some ideas for what sorts of things could be done
There might be some tradeoffs available between memory use, CPU use,
vocabulary size, market size (as restricted by patents) and
well-known-ness --- Walsh transforms might be "good enough" but not
well-investigated, but need less memory, for example.
Maybe I should ask Eric Brewer.
More information about the FoRK