AES E-Library

K. Kim, A. Baijal, B. Ko, S. Lee, I. Hwang, and Y. Kim, "Speech Music Discrimination Using an Ensemble of Biased Classifiers," Paper 9457, (2015 October.). doi: K. Kim, A. Baijal, B. Ko, S. Lee, I. Hwang, and Y. Kim, "Speech Music Discrimination Using an Ensemble of Biased Classifiers," Paper 9457, (2015 October.). doi:
Abstract: In this paper we present a novel framework for real-time speech/music discrimination (SMD). The proposed method improves the overall accuracy of automatically classifying the signals into speech, singing, or instrumental categories. In our work, first, we design several groups of classifiers such that each group’s classification decision is biased towards a certain class of sounds; the bias is induced by training different groups of classifiers on perceptual features extracted at different temporal resolutions. Then, we build our system using an ensemble of these biased classifiers organized in a parallel classification fashion. Last, these ensembles are combined with a weighting scheme, which can be tuned in either forward-weighting or inverse-weighting modes, to provide accurate results in real-time. We show, through extensive experimental evaluations, that the proposed ensemble of biased classifiers framework yields superior performance compared to the baseline approach.

@article{kim2015speech,
author={kim, kibeom and baijal, anant and ko, byeong-seob and lee, sangmoon and hwang, inwoo and kim, youngtae},
journal={journal of the audio engineering society},
title={speech music discrimination using an ensemble of biased classifiers},
year={2015},
volume={},
number={},
pages={},
doi={},
month={october},} @article{kim2015speech,
author={kim, kibeom and baijal, anant and ko, byeong-seob and lee, sangmoon and hwang, inwoo and kim, youngtae},
journal={journal of the audio engineering society},
title={speech music discrimination using an ensemble of biased classifiers},
year={2015},
volume={},
number={},
pages={},
doi={},
month={october},
abstract={in this paper we present a novel framework for real-time speech/music discrimination (smd). the proposed method improves the overall accuracy of automatically classifying the signals into speech, singing, or instrumental categories. in our work, first, we design several groups of classifiers such that each group’s classification decision is biased towards a certain class of sounds; the bias is induced by training different groups of classifiers on perceptual features extracted at different temporal resolutions. then, we build our system using an ensemble of these biased classifiers organized in a parallel classification fashion. last, these ensembles are combined with a weighting scheme, which can be tuned in either forward-weighting or inverse-weighting modes, to provide accurate results in real-time. we show, through extensive experimental evaluations, that the proposed ensemble of biased classifiers framework yields superior performance compared to the baseline approach.},}

TY - paper
TI - Speech Music Discrimination Using an Ensemble of Biased Classifiers
SP - EP -
AU - Kim, Kibeom
AU - Baijal, Anant
AU - Ko, Byeong-Seob
AU - Lee, Sangmoon
AU - Hwang, Inwoo
AU - Kim, Youngtae
PY - 2015
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - October 2015 TY - paper
TI - Speech Music Discrimination Using an Ensemble of Biased Classifiers
SP - EP -
AU - Kim, Kibeom
AU - Baijal, Anant
AU - Ko, Byeong-Seob
AU - Lee, Sangmoon
AU - Hwang, Inwoo
AU - Kim, Youngtae
PY - 2015
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - October 2015
AB - In this paper we present a novel framework for real-time speech/music discrimination (SMD). The proposed method improves the overall accuracy of automatically classifying the signals into speech, singing, or instrumental categories. In our work, first, we design several groups of classifiers such that each group’s classification decision is biased towards a certain class of sounds; the bias is induced by training different groups of classifiers on perceptual features extracted at different temporal resolutions. Then, we build our system using an ensemble of these biased classifiers organized in a parallel classification fashion. Last, these ensembles are combined with a weighting scheme, which can be tuned in either forward-weighting or inverse-weighting modes, to provide accurate results in real-time. We show, through extensive experimental evaluations, that the proposed ensemble of biased classifiers framework yields superior performance compared to the baseline approach.