Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!brutus.cs.uiuc.edu!apple!goofy.apple.com!esmith From: esmith@goofy.apple.com (Eric Smith) Newsgroups: comp.dsp Subject: Re: Speech Recognition help wanted Message-ID: Date: 27 Aug 90 18:14:00 GMT References: <6698.26d02a31@ccvax.ucd.ie> Sender: usenet@Apple.COM Followup-To: comp.dsp Organization: Frobozz Magic Widget Company Lines: 32 In-reply-to: josef@nixpbe.UUCP's message of 23 Aug 90 11:06:08 GMT In article josef@nixpbe.UUCP (Moellers) writes: >In <6698.26d02a31@ccvax.ucd.ie> b_haughey@ccvax.ucd.ie (Brian J Haughey) writes: >>I'm looking for information on speech recognition chips. Could any of you >>give me some pointers on references ? > >A couple of years ago BYTE Magazine (when it was still usefull) had an >article or two by Steve Ciarcia on a speech recognition system based on >a chip made by General Instruments (I think). It was called something >like SP256. The interesting thing about this chip is that it is >advertised as being a speech synthesis chip. The chip you're thinking of is the General Instruments SP-1000, which does LPC (linear predictive coding) analyis and synthesis. GI had application notes for speech recognition. The LPC filter they used was configured differently for analysis than for synthesis, so you can't use the chip to record then play back a sound. It seemed somewhat brain-dead to me. It had the advantage of being cheap enough to put in consumer products, although I'm not sure that anyone ever did so. The BYTE article described a board for the Apple II which used this chip. The board was available from MicroMint (sp?). If anyone wants some of these chips, I've got some I could sell. It probably makes more sense to use a DSP chip for speech recognition now. Motorola has some LPC software on their bulletin board. -- Eric L. Smith Opinions expressed herein do not necessarily reflect those esmith@apple.com of my employer, friends, family, computer, or even me! :-)