Yes. Or else it's a machine learning problem at far side, with speakers organized by, I dunno, geography.

Regardless, the models will need training.

Al Matthews,
AUC Robert W. Woodruff Library

404.978.2057 o
404.769.2617 c

----- Reply message -----
From: "Gary McGath" <[log in to unmask]>
To: "[log in to unmask]" <[log in to unmask]>
Subject: [CODE4LIB] Oral history app and server
Date: Wed, Oct 3, 2012 5:06 pm



Continuing on this part: My friend says that using any existing speech
recognition software won't work at all well for transcribing interviews
with a variety of people. All such software needs to be "trained" to the
speaker's voice.

A possible alternative is for a designated person to train the software
and re-speak it into the speech recognition software.

On 10/3/12 6:22 AM, Gary McGath wrote:
> On 10/2/12 8:44 AM, Paul Orkiszewski wrote:

>> - Processes the audio through speech recognition either in real time or
>> post-interview, and populates the dbase record with rendered text (at
>> whatever level of accuracy)
>
> You could do this piece with Dragon; see this post for some discussion:
>
> http://www.nuance.com/dragon/transcription-solutions/index.htm
>
> A friend of mine is an expert in this area and might be able to answer
> some questions.



--
Gary McGath, Professional Software Developer   http://www.garymcgath.com

-----------------------------------------
**************************************************************************************************
The contents of this email and any attachments are confidential.
They are intended for the named recipient(s) only.
If you have received this email in error please notify the system
manager or  the 
sender immediately and do not disclose the contents to anyone or
make copies.

** IronMail scanned this email for viruses, vandals and malicious
content. **
**************************************************************************************************