[Air-L] voice recognition software for interview transcription?

Alex Halavais alex at halavais.net
Fri Jun 24 10:39:28 PDT 2011


Yes, this seems to be a perennial question for AIR-L. And the answer
has thus far remained the same: automatic transcription just isn't
there yet. If you see some of the transcriptions offered by YouTube,
you'll know why. If you are only looking for automatic tagging of
keywords, there are tools and services available to do this, with the
better ones getting quite expensive, but for accurate transcriptions,
humans are still needed.

I haven't used Mechanical Turk for transcription, but I'd be
interested to know if anyone on the list has...

For reference:

http://www.cs.cmu.edu/~mrmarge/icassp10margem.pdf
http://waxy.org/2008/09/audio_transcription_with_mechanical_turk/
http://chronicle.com/blogs/profhacker/the-reliability-efficiencyaffordability-of-amazons-mechanical-turk/22994
http://castingwords.com/ (They seem to use MTurk, but wrap it for customers.)

Best,

Alex


On Fri, Jun 24, 2011 at 1:09 PM,  <jhuns at vt.edu> wrote:
> I think many people are looking and have been looking for this holy grail.
>  What most people have done when they have a good deal of data is hire a
> transcription service to do it for them
> _______________________________________________
> The Air-L at listserv.aoir.org mailing list
> is provided by the Association of Internet Researchers http://aoir.org
> Subscribe, change options or unsubscribe at:
> http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>
> Join the Association of Internet Researchers:
> http://www.aoir.org/
>



-- 
//
// This email is
// [x] assumed public and may be blogged / forwarded.
// [ ] assumed to be private, please ask before redistributing.
//
// Alexander C. Halavais, ciberflâneur
// http://alex.halavais.net
//



More information about the Air-L mailing list