Rapid learning of subjective preference in equalization

Andrew T. Sabin*, Bryan A Pardo

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

We describe and test an algorithm to rapidly learn a listener's desired equalization curve. First, a sound is modified by a series of equalization curves. After each modification, the listener indicates how well the current sound exemplifies a target sound descriptor (e.g., "warm"). After rating, a weighting function is computed where the weight of each channel (frequency band) is proportional to the slope of the regression line between listener responses and within-channel gain. Listeners report that sounds generated using this function capture their intended meaning of the descriptor. Machine ratings generated by computing the similarity of a given curve to the weighting function are highly correlated to listener responses, and asymptotic performance is reached after only ∼25 listener ratings.

Original languageEnglish (US)
Title of host publicationAudio Engineering Society - 125th Audio Engineering Society Convention 2008
Pages1336-1342
Number of pages7
StatePublished - Dec 1 2008
Event125th Audio Engineering Society Convention 2008 - San Francisco, CA, United States
Duration: Oct 2 2008Oct 5 2008

Publication series

NameAudio Engineering Society - 125th Audio Engineering Society Convention 2008
Volume2

Other

Other125th Audio Engineering Society Convention 2008
CountryUnited States
CitySan Francisco, CA
Period10/2/0810/5/08

ASJC Scopus subject areas

  • Modeling and Simulation
  • Acoustics and Ultrasonics

Fingerprint Dive into the research topics of 'Rapid learning of subjective preference in equalization'. Together they form a unique fingerprint.

Cite this