Re: Echo II sound cards

[anoneds@netscape.net (Ben Yates)]

mjmahon@aol.com (Michael J. Mahon) wrote in message news:<20040327175640.01536.00000215@mb-m15.aol.com>...
> Jay wrote:
> 
> >The company that sold them and sells the IBM equivalent is in California.  I
> >made some phone calls to them about three years ago.  The manual states that
> >there are three ways that the Echo can produce humans speech:  a robotic male
> >voice with unlimited vocabulary, a high-quality female voice with a 4000 word
> >vocabulary, and a third way.  I was interested in learning more about the 3rd
> >option, or how to make my own high quality compressed speech (to have Fred
> >Flintstone say, "Yaba-daba-do!" on my Apple II, for example).  The president
> >and v.p. of the company are still there, but everyone else is a newbie, Apple
> >II-wise.  In the end, they couldn't find the docs, so it was a dead end.
> 
> I contacted them in the early 90's, and had a little bit more luck.
> 
> The "third way" was to send 8-bit digitized sound samples to the
> card and just use it as a D/A converter.  The sample stream mode
> ended when a $00 byte was sent, as I recall.
> 
> I tried the method, but it didn't work on my Echo card.  I suspect
> the problem may have been with the version of the card, but don't
> know for sure.
> 
> It is possible to send LPC coefficients to the card, since that's
> how it makes the phoneme-synthesis sounds (the unlimited
> vocabulary "male" voice) and the pre-computed female voice.
> 
> I'm not aware of any tools they ever provided to create the
> tables of LPC coefficients that make up the female voice
> vocabulary tables or the male voice phoneme tables.
>

And if you _do_ find a tool to create LPC code, please let me know.
I've searched long for it. Come up with completed products that use
it, or specialized software requiring a certain sound board to play it
- although the utilities are supposed to create LPC without the board,
it doesn't look like any LPC to me, nor does it work.

In the meantime, I can use my TI Text-to-Speech disk along with
software to capture the LPC strings to create new words. This isn't
the same as digitally sampling a voice and converting, but it is the
best I can do...

Ben