Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!hookup!usc!howland.reston.ans.net!ix.netcom.com!netcom.com!ebohlman
From: ebohlman@netcom.com (Eric Bohlman)
Subject: Re: Commerical pre-recorded Voice Dictionaries
Message-ID: <ebohlmanD30AD2.8Eu@netcom.com>
Organization: OMS Development
X-Newsreader: TIN [version 1.2 PL1]
References: <Pine.SUN.3.91.950124161525.24411A-100000@svha0001.clipper.ssb.com> <slerner-2501951045200001@slerner.gte.com>
Date: Thu, 26 Jan 1995 09:29:26 GMT
Lines: 15

Sol Lerner (slerner@gte.com) wrote:
: In article
: <Pine.SUN.3.91.950124161525.24411A-100000@svha0001.clipper.ssb.com>, Mark
: Hoffman <mehoffm@clipper.ssb.com> wrote:

: > Does anyone know of any pre-recorded dictionaries of English words
: > in a standard voice format, such as PCM u-law? I am looking for
: > something other than the typical text-to-speech package because I want 
: > to get very realistic sounding sentences and the access does not have
:          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
: Hmmm, that may be hard to do since there would be no way to control for prosody.

Even worse, there's no way to achieve co-articulation, and lack of 
co-articulation makes even a human voice sound robotic.

