Title :
Semi-blind algorithm for joint noise suppression and dereverberation based on higher-order statistics and acoustic model likelihood
Author :
Aprilyanti, Fine D. ; Saruwatari, Hiroshi ; Shikano, Kiyohiro ; Nakamura, Shigenari ; Takatani, Tomoya
Author_Institution :
Grad. Sch. of Inf. Sci., Nara Inst. of Sci. & Technol., Nara, Japan
fDate :
Oct. 29 2013-Nov. 1 2013
Abstract :
In this paper, we propose an automatic optimization scheme of FD-BSE-based joint suppression of noise and late reverberation to improve the speech recognition accuracy for spoken-dialogue system. First, we optimize the parameter of conventional FD-BSE-based method using the assessment of musical noise measured by higher-order statistics and acoustic model likelihood. Next, to maintain the optimum performance of the system, we proposed the switching scheme using the distance information provided by image sensor. The experimental results show that the proposed approach improves the word recognition accuracy.
Keywords :
blind source separation; frequency-domain analysis; image sensors; interactive systems; optimisation; reverberation; signal denoising; speech recognition; FD-BSE-based joint noise suppression; acoustic model likelihood; automatic optimization scheme; dereverberation; distance information; frequency-domain blind signal extraction; higher-order statistics; image sensor; late reverberation; musical noise assessment; semiblind algorithm; speech recognition accuracy; spoken-dialogue system; switching scheme; word recognition accuracy; Accuracy; Joints; Noise; Optimization; Reverberation; Speech; Speech recognition;
Conference_Titel :
Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2013 Asia-Pacific
Conference_Location :
Kaohsiung
DOI :
10.1109/APSIPA.2013.6694161