Return to the archive index

Re: CMU Sphinx Speech Recognition

From: Alex Feinman <>
Date: Thu, 3 Feb 2000 21:40:03 -0500

Charles J Knight wrote:
> > The speech engine itself is 1.2 MB, the whole of ViaVoice/bin is 
> > only 2.2 MB, and the libraries are pretty small.  A single file,
> > accounts for 73 MB.  Presumably this can be shrunk down a lot if you
> 
> Out of curiosity...why are we trying to implement speaker independent
> voice recognition, on a single user machine?
> 
> We need only a small vocabulary for control (not dictation), and for
> only a single user.

Personally, I want a completely voice-controlled machine, and for that
I do need to be able to do dictation, of english or code.

Also, having to redefine your vocabulary (and possibly remunge grammar
rules or whatnot) every time you add new functionality is annoying.

>      -- Chuck Knight

Alex

--
Subcription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to 
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org

+Previous Message in Thread | Next Message in Thread

From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty

Archive created with babymail