DocumentCode :
2787601
Title :
A psychoacoustic spectral subtraction method for noise suppression in automatic speech recognition
Author :
Haque, Serajul ; Togneri, Roberto
Author_Institution :
Sch. of Electr., Electron. & Comput. Eng., Univ. of Western Australia, Crawley, WA, Australia
fYear :
2010
fDate :
14-19 March 2010
Firstpage :
1618
Lastpage :
1621
Abstract :
A time-frequency spectral subtraction method based on the knowledge of several psychoacoustic properties of human perception is presented. These effects are the critical band filtering, synaptic adaptation which also introduces temporal forward masking, equal loudness preemphasis, power law of hearing, and simultaneous masking effect. The perceptual speech and noise is estimated separately by a detailed psychoacoustic non-linear transformation undergoing in the human auditory system. The spectral subtraction using a over-subtraction factor and a spectral floor is measured by a speech recognition front-end using a continuous density HMM recognizer. The method shows reduced residual noise and improved word recognition performance in broadband Gaussian noise conditions compared to conventional spectral subtraction method.
Keywords :
Gaussian noise; acoustic signal processing; hidden Markov models; interference suppression; speech enhancement; speech intelligibility; speech recognition; automatic speech recognition; broadband Gaussian noise; continuous density HMM recognizer; critical band filtering; equal loudness preemphasis; hearing power law; human auditory system; human perception; noise suppression; nonlinear transformation; psychoacoustic spectral subtraction method; simultaneous masking effect; speech recognition front-end; synaptic adaptation; temporal forward masking; word recognition performance; Auditory system; Automatic speech recognition; Density measurement; Filtering; Gaussian noise; Humans; Psychology; Speech enhancement; Speech recognition; Time frequency analysis; Auditory system; acoustic signal processing; speech communication; speech enhancement; speech recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on
Conference_Location :
Dallas, TX
ISSN :
1520-6149
Print_ISBN :
978-1-4244-4295-9
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2010.5494888
Filename :
5494888
Link To Document :
بازگشت