DocumentCode :
2166136
Title :
Feature selection based on Multiple Kernel Learning for single-channel sound source localization using the acoustic transfer function
Author :
Takashima, Ryoichi ; Takiguchi, Tetsuya ; Ariki, Yasuo
Author_Institution :
Graduate School of System Informatics, Kobe University, 1-1 Rokkodai, Nada-ku, 657-8501 Japan
fYear :
2011
fDate :
22-27 May 2011
Firstpage :
2696
Lastpage :
2699
Abstract :
This paper presents a sound source (talker) localization method using only a single microphone. In our previous work [1], we discussed the single-channel sound source localization method, where the acoustic transfer function from a user´s position is estimated by using a Hidden Markov Model (HMM) of clean speech in the cepstral domain. In this paper, each cepstral dimension of the acoustic transfer function is newly selected in order to select the cepstral dimensions having information that is useful for classifying the user´s position. Then, we propose a feature selection method for the cepstral parameter using Multiple Kernel Learning (MKL) to define the base kernels for each cepstral dimension (scalar) of the acoustic transfer function. The user´s position is trained and classified by Support Vector Machine (SVM). The effectiveness of this method has been confirmed by sound source (talker) localization experiments performed in a room environment.
Keywords :
Cepstral analysis; Hidden Markov models; Kernel; Speech; Support vector machines; Transfer functions; Multiple Kernel Learning; feature selection; maximum likelihood; single channel; talker localization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
Conference_Location :
Prague, Czech Republic
ISSN :
1520-6149
Print_ISBN :
978-1-4577-0538-0
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2011.5947041
Filename :
5947041
Link To Document :
بازگشت