On Sat, 16 Dec 2006 12:48:15 +0200, [log in to unmask] wrote:
...
>Having tie-clip microphones on all participants this could be extended to
>group interviews if there were software to support several microphones.
>With just one audio-stream one would need automatic speaker-recognition,
With Transcriber you can load several stereo (or mono) recordings at the
same time. So if you were doing a focus group or recording a meeting you
could use several cheap digital recorders with a tie clip for each speaker
allowing each speaker to have their own channel. You might have to play
around with each stereo track in an audio editor to get them all to start at
exactly the same point. Anyway, multiple tracks can be loaded at the same
time. When you transcribe you switch back and forth between tracks and
channels. On an individual stereo track you can listen to both channels at
ounce or left or right.
I think some of this functionality was added based on modifications made to
the program by a group in Berkeley that had money from DARPA, IBM and others
to work on a speech recognition system for meetings. See
http://www.icsi.berkeley.edu/Speech/mr/channeltrans.html
|