[comp.sys.ibm.pc] Voice recognition/speech synthesis products

doc@holin.ATT.COM (David Mundhenk) (12/30/88)

I posted this several weeks ago (I think) with no response, so I want 
to try one more time before I dump a lot of money into this.
Please pardon the bandwidth.

I have been very interested in speech synthesis and voice recognition
for quite a while (had a nice speech synthesizer for my Commodore 64,
rest it's soul). November PC Tech Journal has an ad in the back from
"COVOX,Inc." (TM?) listing "Speech Products" including "Synthesizer-
$79.95", "Digitizer-$89.95" and "Voice Recognition-$49.95".

Anyone familiar with these products? I got money for these for Christmas
but don't want anyone to waste money on junk.  :v(
The prices sound almost too good to be true....

I commonly use two computers/terminals at once, and it would be *nice*
to talk to one while I type on another. 8-]

I also want to investigate the possibility of developing applications
for the handicapped (or the lazy).

If no one has heard of these products I will probably try them and
let you know what I think.

-Dave

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
EMAIL: ...!att!holin!doc  | "I can't complain but |   /^,
VOICE: (201)-949-5308     |  sometimes I still do"|  /  } _, , , __
#include <std.disclaimer> |  - Joe Walsh          | /_./ (_l |/ <~_
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

geek@mit-amt (Chris Schmandt) (12/30/88)

In article <300@holin.ATT.COM> doc@holin.ATT.COM (David Mundhenk) writes:
>
>I have been very interested in speech synthesis and voice recognition
>for quite a while (had a nice speech synthesizer for my Commodore 64,
>rest it's soul). November PC Tech Journal has an ad in the back from
>"COVOX,Inc." (TM?) listing "Speech Products" including "Synthesizer-
>$79.95", "Digitizer-$89.95" and "Voice Recognition-$49.95".
>
>Anyone familiar with these products? I got money for these for Christmas
>but don't want anyone to waste money on junk.  :v(
>The prices sound almost too good to be true....
>
>I commonly use two computers/terminals at once, and it would be *nice*
>to talk to one while I type on another. 8-]

I would not use the term *junk* but for these prices you're not going
to get something that really does what you think it will unless you've
got some experience dealing with speech technologies.  These will be
toys; hopefully fun toys, but nothing that will really change the way
you think about your computers.  The synthesized speech will be
hard to understand; you may in time become quite proficient at
understanding it, but others will have a hard time.  Speech 
recognition with devices costing 1 to 2 orders of magnitude more
is still not good enough that you'd really want to replace the
keyboard.  (voice might be useful for switching one keyboard
between 2 screens, though).

Covox has been around for a while and their stuff works.  It's not
great.  Mind you, the $5,000 equivalents arent' that great either.

In short: speech is a tough world to work in.  Get the stuff and 
start to learn about it, but don't expect miracles!

chris

nelson@sun.soe.clarkson.edu (Russ Nelson) (12/30/88)

In article <3442@mit-amt> geek@mit-amt (Chris Schmandt) writes:

   In short: speech is a tough world to work in.  Get the stuff and 
   start to learn about it, but don't expect miracles!

Chris is right, most of the speech stuff out there are toys.  So, if you
*really* want to play, go to your local Radio Shack store and check out
their speaker-independent speech recognition chip.  It's a fairly clever
hack that relies on the ease of recognizing the fricative.  So, they
arrange their vocabulary so that it's either (voiced), (fricative),
(voiced) (fricative), or (fricative) (voiced).  I think it's something like
"Go" (or "No"), "Stop" (or "Yes"), "Turn Left", or "Right Turn".

Anyway, it's real cheap and looks like it might work, so check it out.
--
--russ (nelson@clutx [.bitnet | .clarkson.edu])
"I saved the whales!" - Rebecca L. Nelson, 3.5 years old, on receiving her
Christmas present of a whale "adoption" certificate.  Bless her liberal heart.

hundt@eagles.rutgers.edu (Hundt) (12/30/88)

|"COVOX,Inc." (TM?) listing "Speech Products" including "Synthesizer-
|$79.95", "Digitizer-$89.95" and "Voice Recognition-$49.95".

Well I'd check first if they aren't using the National chipset; you
could buy this at Radio Shack and build your own for $25.  (if you're
handy at hardware that is)

I'm pessimistic about speech recognition products; basically it's
turned into a software problem that hasn't been solved yet.  They've
gotten as far as a toy (called "voyager" or something) that recognizes
"yes" and "no" and that's about all.  IBM has been working on a device
for many years to take speech (eg. dictation) and turn it into data
(eg. wordprocessor file).  I don't see it on the market yet, so I
don't know, but I'd kinda doubt the effectiveness of the $49.95 speech
recognizer.

Yours doubtfully,
Tom

hundt@occlusal.rutgers.edu:Thomas M. Hundt:201/247-6723(H),932-5843(Lab)
-- 
RRRRRR    Thomas M. Hundt
 RR  RR   Gradual Student --- Electrical & Computer Engineering
 RR  RR   Rutgers University
 RRRRR    New Brunswick NJ
 RR  RR   hundt@occlusal.rutgers.edu
RRR  RRR  Famous last words: "The virus ate it."

SHENF@kcgl1.eng.ohio-state.edu (Zhiyong Shen) (12/30/88)

  I saw the FOX TV-network program "Beyond Tomorrow" a month back.  In the
program they introduced the Italian Computer Maker Olivetti which has
successfully developed a voice recognition program plus hardware.  The program
converts spoken words into ascii characters.  Therefore the technology seems
to be more advanced than what the discussions in this newsgroup suggest.  Of
course it might be too expensive for hackers right now.

rgr@m10ux.UUCP (Duke Robillard) (12/30/88)

In article <300@holin.ATT.COM> doc@holin.ATT.COM (David Mundhenk) writes:
>I have been very interested in speech synthesis and voice recognition
>.... November PC Tech Journal has an ad in the back from
>"COVOX,Inc." (TM?) listing "Speech Products" including "Synthesizer-
>$79.95", "Digitizer-$89.95" and "Voice Recognition-$49.95".

I bought all the Covox stuff (mostly cause it sounded cool and
was cheap.  And I love I/O stuff.  anyone got a track-ball they
wanna unload?  Or any footswitchs?  Has anyone bought that funky
flight simulator plane control thing that clamps on your desk?)

When I got it, it was in two parts: a speech synth and a
digitizer/voice recog thing.  Looks like they've unbundled the
voice recog. software.

The voice synthesizer is pretty good.  It just a 8 bit DAC in 
a module that plugs into your parallel port (without hurting
your printer) and a little speaker.  There's a device driver
for putting out digitized voice files (you can do this from
inside a program, so all your error message can be vocal)
and a third-party program that'll read text. Each bunch of
digitized info that goes out has to be less than 64K, which 
is only a couple of secs (8 secs, I think, at the worst
resolution/sampling rate) but you can send one after another
without much gap between 'em.  The text reader has that
computer voice, but is pretty good.  It'll get words wrong
occasionally, but not too bad.

The digitizer is a 1/2 size  card with an 8 bit ADC and a 
microphone plug.  It's good for generating files for the
synthesizer to speak.  From what I can see, the voice 
recognition software is kinda lame.  It's easy to use,
but not particularly good.  It's definitely speaker-dependent
(i.e. recognizing one individual doesn't help in recognizing
anyone else) and you really have to speak the same way you did 
when you were training it.  This is also a device driver. To
use it you send a "train" command and some training data for the
each word you want to recognize (64 max).  Then you pass it
other data and it returns the number of the word that matched,
or some error code.

TI has just come out with a new voice recog. chip (I read
about it in EE Times yesterday) so maybe you should wait on
that part.  There are gonna be speaker independent voice
activated toys soon, so can a good PC board be far away?

There's a cool digitized voice file editor that comes with
both the synth and the digitizer.  This is useful for 
doctoring recordings (:->)  I can see it now:  
"Did you kill him?  speak into my computerized recorder."  
"I did not!"
(edit, edit) REPLAY: 
"Did you kill him?  speak into my computerized recorder."  
"I did!"


-- 
|  Duke Robillard                                                        |
|                                  |                                     |
|  ARPA:     rgr@m10ux.att.com     | UUCP:     {backbone!}att!m10ux!rgr  |
|  or maybe: m10ux!rgr@att.att.com | BITNET:   rgr%m10ux.uucp@psuvax1 (?)|

webb@leadsv.UUCP (J.J. Webb) (12/31/88)

There is a speech recognition board made by Interpath, Inc., in Santa 
Clara. The board sells for 4-5 hundred dollars and has a vocabulary of about 
500 words. I've seen demos of it being used with DOS, and with Lotus, DBase 
and some other applications ... it was very impressive. The phone
number is 408/727-4559. Ask for John McAfee, he can tell you whole
bunches about speech rec.

There is also a board made by (I think the company's name is) Votrax. 
I saw a demo of it as well ... it was less impressive ... fewer words,
harder to train, higher cost, etc.   They weren't as impressive as
Interpath so I dismissed them pretty fast.

But speech rec. is in its infancy. And speech synth. is real tough.
Companies working on speech synth for the handicapped include IBM and
Computer Curriculum Corporation. A great deal of research has been
done at Standford Univ. as well. CCC is in Palo Alto. I don't have
the phone number handy but could dig it up fairly quickly if anyone
is interested. 

                      ....jjwebb....

BTW ... I worked for CCC 10 years ago and worked with John McAfee 5
years ago, but I have no professional connection with either CCC or 
Interpath nowadays. 

network@hgcvax.uucp (craig chaiken) (12/31/88)

In article <300@holin.ATT.COM>, doc@holin.ATT.COM (David Mundhenk) writes:
> November PC Tech Journal has an ad in the back from
> "COVOX,Inc." (TM?) listing "Speech Products" including "Synthesizer-
> $79.95", "Digitizer-$89.95" and "Voice Recognition-$49.95".
> 
> Anyone familiar with these products? I got money for these for Christmas
> but don't want anyone to waste money on junk.  :v(
> The prices sound almost too good to be true....
>
I purchased all three COVOX products several months ago.  The best
classification for these products is "experimental."  That is, the products
function, but are not sophisticated enough to handle real applications.  The
hardware simply consists of an analog to digital convertor for sound input,
and a digital to analog convertor for sound output.  Voice recognition and
synthesis are performed totally in software.  The voice recognition is
very poor; the software is supposed to recognize 64 words, but I was never
able to achieve even 50% reliability.  The voice synthesis is very slow
on even 80386 machines.  It is hardly understandable on XT's, but is
quite realistic on my 80386 machine.

This is not to say that I do not recommend the products.  The COVOX
Speech Products are the least expensive hardware around for dabbling
in voice I/O, and for the price they perform pretty good.  However,
do not bother integrating them into serious applications, because
they just aren't sophisticated enough.

Craig Chaiken
Hartford Graduate Center
Computing Services