|
Carlos,
I have done this job when I created 2-, 4-, and
8-talker babble noise for testing cocktail-party performance,
using the IEEE sentence corpus. Unfortunately however, I added the
output of the talkers, i.e., the babble is single-channel. But I
recommend that corpus: it is fairly extensive and when you play
the sentences back-to-back starting at random points of the list,
you have the independent tracks you need.
Best,
Pierre
On 1/17/17 5:54 PM, Benitez, Carlos R
wrote:
I am Carlos Benítez a PhD student at Vanderbilt
University (USA). I’m looking for a four english-talker
babble noise with independent tracks for each of the
talkers, so I can delivered the babble noise
simultaneously from four different loudspeakers (One
talker - one speaker) for one of my experiments.
I was wondering if someone on this list would have
developed a stimuli like that before and whether they
would be willing to share it with me. Please if you have
such an stimuli on your corpus contact me at
carlos.r.benitez@xxxxxxxxxxxxxx.
I would really appreciate your help!
Thanks in advance,
Carlos Benitez
|