[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: pitch neurons

Hello List,

    From my experience in speech I fully agree
with Eli's remark. I would even argue that the
very notion of pitch differs according to the
abstraction level considered; in my view speech
processing implies several levels of abstraction
(5 or 6) which correspond to as many steps in time
resolution, ranging from some 15 ms to several
hundreds of ms at the cognitive level. The F0
measurement that appears at low level is close to
a physical property of the signal. It is not
defined during the unvoiced segments. However, at
higher levels (syllable-size level, word-size
level, phrase-size level...) we perceive
continuous, smooth pitch properties, loosely
related to the low-level F0. We attach different
meanings to them: lexical stress, phrase accent,
demarcation between linguistic units,
interrogation or assertion, speaker's gender,
acuteness of the speaker in his/her gender range,
emotional state, listener's distance, etc. As Eli
points out, the perception of these properties is
highly dependent on cognitive and contextual
factors. Maybe what we call "pitch" actually
refers to a set of different notions that should
be defined and processed differently according to
their level of abstraction...

Israel Nelken wrote:

>       I would like however to raise another issue: why is >there such a pressure to assume low-level representation (i.e. >subcortical) of pitch? After all, pitch is a pretty high-level >property of sounds, it is invariant to many features of the >physical structure of the sounds, and it is affected by all >kinds of high-level cognitive effects (e.g. capture of >harmonics in streaming conditions). All of these would suggest >to me that whatever is responsible for the pure representation >of pitch (independent of the physical structure of the sounds) >is rather high-level, rather than low-level.
>        Eli

Jean-Sylvain Liénard
    BP 133  91403 Orsay Cedex
    tél      33 1 69 85 81 13
    fax      33 1 69 85 80 88
    E-mail   jean-sylvain.lienard@limsi.fr