Co-channel speech separation using frequency bin non-linear adaptive filtering

Author

Gu, Y.H. ; Van Bokhoven, W. M G

Author_Institution

Dept. of Electr. Eng., Eindhoven Univ. of Technol., Netherlands

fYear

1991

fDate

14-17 Apr 1991

Firstpage

949

Abstract

An approach to speech separation via frequency bin nonlinear adaptive filtering is proposed. The algorithm is proved effective for speech intelligibility enhancement over a TIR (target-to-interference energy ratio) range between 0 dB and ±12 dB. Informal listening tests and spectrogram comparisons are discussed. In addition, a robust multi-pitch contour estimation via HMMs (hidden Markov models) is investigated. This HMM pitch estimator is combined with a pseudo-perceptual pitch estimator in order to simultaneously estimate two speakers´ pitch from summed signals

Keywords

Markov processes; adaptive filters; parameter estimation; speech analysis and processing; speech intelligibility; HMM pitch estimator; cochannel speech separation; frequency bin nonlinear adaptive filtering; hidden Markov models; listening tests; multi-pitch contour estimation; pseudo-perceptual pitch estimator; spectrogram comparisons; speech intelligibility enhancement; Adaptive filters; Auditory system; Frequency; Hidden Markov models; Interference; Robustness; Signal processing; Signal processing algorithms; Speech enhancement; Speech processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on

Conference_Location

Toronto, Ont.

ISSN

1520-6149

Print_ISBN

0-7803-0003-3

Type

conf

DOI

10.1109/ICASSP.1991.150497

Filename

150497