Combining Sub-bands SNR on Cochlear Model for Voice Activity Detection

Author

Liu, Qibo ; Liu, Yi ; Li, Yanjie

Author_Institution

Dept. of Auto Control, HIT, Shenzhen, China

fYear

2010

fDate

28-30 Dec. 2010

Firstpage

319

Lastpage

322

Abstract

In this paper, we proposed a novel approach to combine sub-bands SNR for Voice Activity Detection. In the proposed algorithm, a nonlinear method based on cochlear model is used to divide sub-bands. In each sub-band, two Order Statistic Filters are used to estimate the signal SNR. Through the use of the above methods, the sub-bands SNR is combined by a linear dicriminant function calculated under the MSE criterion. The effectiveness of proposed method has been evaluated on RASC863 corpus. It is shown that the proposed algorithm has more robust ability against the common VAD methods, and the non-speech hit rate is significant improved under the proposed algorithm.ε

Keywords

mean square error methods; nonlinear filters; speech recognition; speech synthesis; MSE criterion; RASC863 corpus; VAD methods; cochlear model; linear dicriminant function; nonlinear method; nonspeech hit rate; order statistic filters; signal SNR estimation; subbands SNR; voice activity detection; Artificial neural networks; Databases; Encoding; Signal to noise ratio; Speech; Speech processing; Speech recognition; Discriminant function; MSE; cochlear model; non-speech hit rate; oder statistic filters; sub-bands SNR combination; voice activity detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Asian Language Processing (IALP), 2010 International Conference on

Conference_Location

Harbin

Print_ISBN

978-1-4244-9063-9

Type

conf

DOI

10.1109/IALP.2010.18

Filename

5681580