Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!amdcad!ames!ucbcad!ucbvax!mit-amt.UUCP!jrd
From: jrd@mit-amt.UUCP (Jim Davis)
Newsgroups: comp.dcom.telecom
Subject: Re: Phone Interfaces
Message-ID: <1587@mit-amt.MEDIA.MIT.EDU>
Date: Mon, 28-Sep-87 13:57:59 EDT
Article-I.D.: mit-amt.1587
Posted: Mon Sep 28 13:57:59 1987
Date-Received: Wed, 30-Sep-87 04:30:37 EDT
References: <8709152213.AA03885@pyrnj.uucp> <228@casemo.UUCP>
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: mit-amt!jrd@media-lab.MEDIA.MIT.EDU (Jim Davis)
Organization: MIT Media Lab, Cambridge MA
Lines: 32
Approved: telecom@buit1.bu.edu


In article <228@casemo.UUCP> brian@casemo.UUCP (Brian Cuthie) writes:
>
>In article <8709152213.AA03885@pyrnj.uucp>, pyrnj!romain@RUTGERS.EDU (Romain Kang) writes:
>...
>> They call it a voice response system.  In brief, it can answer the
>> phone, transmit DTMF codes to a computer via an RS-232 port as ASCII
>> digits, receive ASCII text via the same port and send it back down the
>> phone line as high quality synthesized speech....
>
> ...>You have just described about $25 worth of hardware and a 
> few days software.

Perhaps Brian did not notice the mention of synthetic speech.  Synthesizing
speech of high quality is a demanding task for both software and hardware,
and there are no cheap solutions.  Notice also that the input is ASCII text.
This machine reads aloud.  

The current (Sept) issue of the Journal of the Acoustical Society of America
(JASA) contains a fine survey of the state of this art, written by Dennis
Klatt, the researcher most responsible for the synthesizer from which both
both the Prose product mentioned above and the Digital Equipment Corp
DecTalk derive.


appears in
  To be sure, there are cheaper synthesizers


-- 
ARPA:jrd@media-lab.media.mit.edu
Phone: (617)-253-0360