Re: Intelligibility of reversed speech, Why?

[=?X-UNKNOWN?Q?T=F3th_L=E1szl=F3?= <tothl@xxxxxxxxxxxxxxx>]

On Thu, 25 Jan 2001, Ward Drennan wrote:

> > To put it another way: how can you tell that something you removed from
> > the signal was or wasn't a cue?
>
> Ah, that makes it more clear.
> If you looked at intelligibility as a function of signal to noise ratio
> and the functions didn't change, then whatever was removed was not a
> cue. Agreed?

Yes. My only extension would be now that SNR is not the only thing that
can vary but there are many other possibilities like short-band noise,
non-stationary noise, band-pass filtering, etc., etc. These all belong to
what I meant by "all possible conditions".

               Laszlo Toth
        Hungarian Academy of Sciences         *
  Research Group on Artificial Intelligence   *   "Experience is what you
     e-mail: tothl@inf.u-szeged.hu            *    gain when you expected
     http://www.inf.u-szeged.hu/~tothl        *    something else"