Re: FW: Intelligibility of reversed speech, Why?

[John Hershey <jhershey@xxxxxxxxxxxxxxx>]

It's not so much about the properties of hearing as it is about the
properties of speech.
Basically the speech signal is relatively stationary in very short time
windows.  Other signals, such as rapid bird songs for instance, might be
unintelligible under 50ms windowed time-reversal.

----- Original Message -----
From: "Yadong Wang" <ydwang@ELE.URI.EDU>
To: <AUDITORY@LISTS.MCGILL.CA>
Sent: Thursday, January 25, 2001 8:18 AM
Subject: FW: Intelligibility of reversed speech, Why?


Indeed, this is an excellent question, which - strangely, was not addressed
at all by the authors.
Maybe this is due to a poor sensitivity to temporal-envelope phase changes
at high amplitude-modulation rates. Dau (1996) showed that we are
sensitive to phase changes for low-sine modulation
rates only (< 10 Hz); Patterson and coll. (Patterson, 1994; Akeroyd &
Patterson (1997) also showed that sensitivity to envelope time-reversal
degraded at high envelope rates.



______________________
Christian Lorenzi
Equipe Perception Auditive
Laboratoire de Psychologie Experimentale
UMR 8581 CNRS-Université Paris V
Institut de Psychologie
Universite Paris V
71, Avenue Edouard Vaillant
92774 Boulogne-Billancourt Cédex
France

phone: + 33 1 55.20.57.34
Fax:   + 33 1 55.20.58.54