Rapid Learning of Subjective Preference in Equalization
×
Cite This
Citation & Abstract
A. Sabin, and B. Pardo, "Rapid Learning of Subjective Preference in Equalization," Paper 7581, (2008 October.). doi:
A. Sabin, and B. Pardo, "Rapid Learning of Subjective Preference in Equalization," Paper 7581, (2008 October.). doi:
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.
@article{sabin2008rapid,
author={sabin, andrew and pardo, bryan},
journal={journal of the audio engineering society},
title={rapid learning of subjective preference in equalization},
year={2008},
volume={},
number={},
pages={},
doi={},
month={october},}
@article{sabin2008rapid,
author={sabin, andrew and pardo, bryan},
journal={journal of the audio engineering society},
title={rapid learning of subjective preference in equalization},
year={2008},
volume={},
number={},
pages={},
doi={},
month={october},
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.},}
TY - paper
TI - Rapid Learning of Subjective Preference in Equalization
SP -
EP -
AU - Sabin, Andrew
AU - Pardo, Bryan
PY - 2008
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - October 2008
TY - paper
TI - Rapid Learning of Subjective Preference in Equalization
SP -
EP -
AU - Sabin, Andrew
AU - Pardo, Bryan
PY - 2008
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - October 2008
AB - 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.
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.
Authors:
Sabin, Andrew; Pardo, Bryan
Affiliation:
Northwestern University, Department of Communication Sciences and Disorders
AES Convention:
125 (October 2008)
Paper Number:
7581
Publication Date:
October 1, 2008Import into BibTeX
Subject:
Listening Tests & Psychoacoustics
Permalink:
http://www.aes.org/e-lib/browse.cfm?elib=14733