Hi Beth
I am using a voice recognition package. Dragon Naturally Speaking
Preferred. It claims to be able to support multiple users though each one
needs to be trained separately and each user needs to log in as themselves
to access their own unique speech files. I've never used this function
though.
I'm using it for transcribing, I listen to the tape and speak what is said
into the microphone. Its performance is variable sometimes its very good
and sometimes dreadful, making many many mistakes. Then I need to go
through the tape again just to remove all the surreal errors.
The package learns the more you use it, every time you use it it asks you
to save your speech files, you mustn't do this if you've had a bad session,
or have not had time to go through the proper procedure for correcting
mistakes, because it will learn to make more mistakes.
Another thing is that you need a fast PC, I used it on a 250mhtz PC and it
was useless so I moved it to a 450mhtz and its ok though occasionally it
pauses for a long time. I would suggest using the fastest PC you can get
your hands on.
All in all I think its just about worth while, but you have to speak like a
newscaster and you must never have a sore throat.
good luck
Norman Urquía
>Hi all,
>
>I am interested in whether anyone who is on the list has used voice
>recognition software to transcribe interviews for qualitative data analysis.
>I understand from someone who uses the software that some of the software
>can be "trained" to listen to multiple voices, and some only recognizes one
>user's voice. Does anyone know any more information about this? Are you
>aware of softare that can recognize multiple voices?
>
>Any information would be appreciated!
>
>Beth Sperber Richie, Ph.D.
>(301) 523-8882
>
>
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|