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

Re: speech/music

Malcolm Slaney wrote:
> At 1:17 AM -0800 3/31/98, Sue Johnson wrote:
> >I think there must be some way the brain splits up (deconvolves) the
> >signal before applying a speech recogniser.
> No!
> Who says the brain operates in only a bottom-up manner?

Wait. You are arguing that there is some top-down processing, not
that there is no bottom-up processing. Just because there might be
top-down, doesn't mean we need to give up bottom-up. Sue is describing
a clear bottom-up process. I think she has a point. You are describing
something else, that  might require some top-down. This is a bit of a
non sequitur it seems to me.

> The best counter-example I know of is a song by Miriam Makeba where she
> sings in an african click language.  To my non-african ears, a click in the
> middle of speech is heard as speech, but when the same sound is accompanied
> by music it is heard as a drum beat.   An ambiguous sound changes it
> grouping based on the context.  A similar argument can be made about
> sine-wave speech--it has both speech-like and tone-like components.
> Perhaps an even better example is the McGurk Effect-- Wow, a low-level
> auditory decision changes based on visual input.  Certainly the visual
> system isn't connected at a low level to the auditory system.

How do you know? I would expect they are connected at least by the
auditory cortex AC (this is known by now, no? Somebody who knows, help us
here.) The AC is BEFORE speech is recognized, because there is still
a tonotopic representation. That would still be in the bottom up realm,
don't you agree?

> Some
> information *must* be travelling top-down.  It's expectation driven.

While this is true, but it does not rule out bottom-up processing.

> - Malcolm

Jont B. Allen, Room E161; AT&T Labs-Research 180 Park AV.; Florham Park NJ 07932
973/360-8545voice, x8092fax; http://www.research.att.com/info/jba;