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

David Huron's model



RE DAVID HURON'S SCENE ANALYSIS MODEL:

I THINK THAT BASING A WHOLE SCENE ANALYSIS SYSTEM ON PITCH IS NOT
THE RIGHT THING TO DO IN THE GENERAL CASE.  PITCH SHOULD BE ONLY
ONE OF THE TIME x FREQUENCY ANALYSES.  PITCH ITSELF COULD
SOMETIMES DEPEND ON SEQUENTIAL GROUPING, FOR EXAMPLE
THE GROUPING OF PRIOR WITH CURRENT SPECTRAL COMPONENTS WHICH
MIGHT REMOVE SOME OF THE COMPONENTS FROM THE PITCH ANALYSIS.
I WOULD LIKE TO SEE A NUMBER OF ANALYSES DONE FIRST, ONE OF WHICH
WOULD BE "PROBABLE FUNDAMENTALS" (NOT PITCH, WHICH IS A SUBJECTIVE
DIMENSION).  PITCH WOULD COME OUT OF THE ANALYSIS AS WOULD STREAMS,
PERCEIVED LOCATION, TIMBRE DESCRIPTIONS, ETC.

THERE IS A PROBLEM HERE OF CONFUSING ACOUSTIC AND AUDITORY VARIABLES.
THE LATTER EMERGE FROM THE AUDITORY SCENE ANALYSIS, WHILE THE FORMER
ARE USED AS CLUES.

- Al Bregman