Return to the archive index

Re: Timeleine'Puter (was Small Storage Devices

From: "R. Paul McCarty" <>
Date: Mon, 04 May 1998 17:39:05 +0000

So you're thinking of basically a generic DSP card that can be setup to
convert A/D, D/A for whatever you want.  This might work, but it means
text to speech would have to be coded by hand (or at least somewhat if
you take advantages of builtin conversion on chip). I think I found a
couple of speech synthesizer chips that had builtin a/d d/a functions so
it might be possible to find a chip that has the speech synthesis
hardwired.  I was thinking more along the lines that I send text to an
off-board chip and it generates speech (as best it can).

How big is the kit when assembled?

-Paul

Michael Van Donselaar wrote:
> 
> On Mon, 04 May 1998 13:17:24 +0000,  "R. Paul McCarty" <> wrote:
> 
> >I'd really like to add audio to my wearable to do at least text to
> >speech, but I'm hesitant to add yet another card to the pc/104 stack, or
> >add the serial port driven doubleTalk box.  Has anyone thought about
> >using an off the shelf speech synthesizer chip? I've seen single chips
> >at radio shack that do text to speech, and although the quality of the
> >output may be pretty bad, it would be a way to add speech capability to
> >my wearable without adding alot of additional electronics and boards.  I
> >did a quick search and there are lots of vendors selling single chip
> >speech synthesizers, so I imagine some of them are pretty easy to build
> >serial interfaces to.
> >
> >Any suggestions for chips? and whether this is relatively easy to do?
> >
> >-Paul
> 
> How about a DSP solution, like the Analog Devices 2181.  I'm planning on getting an EZ-Kit to
> investigate this.
> 
> My plan is to use an outboard DSP peripheral connected via a serial or parallel port.  This will
> allow speech input and output with lower CPU overhead.
> 
> For speech input the DSP periph samples the mic input and sends feature vectors (Mel scaled
> cepstrum, etc) to the wearable.  Higher level analysis (HMM/neural net etc) takes place on the
> wearable)
> 
> For speech output, the wearable sends low bandwidth data (LPC-10, GSM, maybe even MP3) and drives a
> speaker/headphone.
> 
> Any advice/ideas?

-- 
R. Paul McCarty / DARS Coordinator /  / x52059
317 Lattimore Hall, University of Rochester, Rochester, NY 14627
Computers don't make errors; what they do, they do on purpose.-Dale/KOTH

Previous Message in Thread | Next Message in Thread

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

Archive created with babymail