I'd say at least five years. Here's the bleeding edge:
http://www.icsi.berkeley.edu/Speech/mr/
----- Original Message -----
From: "Richard Gaskin" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, May 02, 2003 7:13 PM
Subject: Re: Systematic Digital recording and transcribing
Sarah Chung wrote:
> Does anyone know how can I record the interview that goes into a system
> which will produce the word processing document (interview data after
> systematic transcribing). Is there anything like that in the market? I
> assume this method will save time and cost.
A Mr. Richmond pointed out, the challenge is that current voice recognition
software must be trained to recognize a particular voice. Most interviews
have at least two voices, and a study will often contain dozens of
interviewees, each with their own dialect, inflection, tone, etc. I think
we're still about five years from commercial deployment of such a tool.
Maybe the cheapest option is an intern? ;)
--
Richard Gaskin
Lead Programmer, HyperRESEARCH
______________________________
[log in to unmask]
|