W3C home > Mailing lists > Public > www-voice@w3.org > October to December 2000

Re: free-form Voice Input

From: Liang Shen <liang@array.ca>
Date: Mon, 2 Oct 2000 09:06:29 -0400 (EDT)
Message-Id: <200010021306.JAA20601@fuji.array.ca>
To: Giles Cope <giles.cope@xtempus.com>
CC: "'James P. Salsman'" <bovik@best.com>, liang@array.ca, www-voice@w3.org
First, I believe transcribe engine is different from speech recognition
engine. There are some vendors offering transcribe engine, like L&H and
IBM.

I don't think there is any VoiceXML servers support transcribe extension
today. However, we, (VoiceGenie Technologies Inc.), may be interested in
adding such capability if there is enough market demands.

By the way, we do have a pure VoiceXML telephony server.

Cheers,

Liang
liang@voicegenie.com

> >What if Giles wanted to transcribe an out-of-vocabulary word?
> >
> 
> It is my understanding that the proposed transcribe tag in the
> VoiceXML 1.0 spec will access the dictation grammar of the automatic
> speech recognition, and so should have a large enough grammar for
> most applications. For specific applications the ASR could be augmented
> with industry specific grammars (E.g. www.speech.phillips.com).
> 
> The question on my mind now is, which VoiceXML servers support the
> transcribe extension if any?
> 
> Giles Cope
> V-Commerce Developer
> 
Received on Monday, 2 October 2000 09:06:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 October 2006 12:48:53 GMT