Re: AUDITORY Digest - 10 May 2001 (#2001-88) (Ward Drennan )

Subject: Re: AUDITORY Digest - 10 May 2001 (#2001-88)
From:    Ward Drennan  <ward(at)IHR.GLA.AC.UK>
Date:    Fri, 11 May 2001 23:22:29 +0100

> perhaps it [streaming] should be considered a special form of grouping, namely > grouping-by-source. > Yes, I'd agree that streaming is a special kind of grouping, operating in the temporal dimension. It could be however grouping by frequency proximity, loudness level (a sequence of quiet tones may segregate from a sequence of loud tones), or timbre. In a cocktail party, you could stream a voice by its source, using all the available grouping cues, reinforced by repetition over time (comodulation over time, consistent voice timbre and roughly constant voiced fundamental, consistent spatial cues over time, etc.) Ward Drennan

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University