DocumentCode :
2076047
Title :
Gender effect cannonicalization for Bangla ASR
Author :
Asfak-Ur-Rahman, M. ; Kotwal, Mohammed Rokibul Alam ; Hassan, Foyzul ; Ahmmed, S. ; Huda, Mohammad Nurul
Author_Institution :
Dept. of Comput. Sci. & Eng., United Int. Univ., Dhaka, Bangladesh
fYear :
2012
fDate :
22-24 Dec. 2012
Firstpage :
179
Lastpage :
184
Abstract :
This paper presents a Bangla (widely used as Bengali) automatic speech recognition system (ASR) by suppressing gender effects. Gender characteristic plays an important role on the performance of ASR. If there is a suppression process that represses the decrease of differences in acoustic-likelihood among categories resulted from gender factors, a robust ASR system can be realized. In the proposed method, we have designed a new ASR incorporating the Local Features (LFs) instead of standard mel frequency cepstral coefficients (MFCCs) as an acoustic feature for Bangla by suppressing the gender effects, which embeds three HMM-based classifiers for corresponding male, female and geneder-independent (GI) characteristics. In the experiments on Bangla speech database prepared by us, the proposed system has achieved a significant improvement of word correct rates (WCRs), word accuracies (WAs) and sentence correct rates (SCRs) in comparison with the method that incorporates Standard MFCCs.
Keywords :
gender issues; hidden Markov models; natural language processing; signal classification; speech recognition; Bangla ASR; Bangla speech database; HMM-based classifier; MFCC; Mel frequency cepstral coefficient; SCR; WA; WCR; acoustic feature; acoustic likelihood; automatic speech recognition system; gender characteristic; gender effect cannonicalization; gender effect sppression; hidden Markov model; local feature; sentence correct rate; word accuracy; word correct rate; acoustic model; automatic speech recognition; gender effects suppression; hidden Markov model;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Information Technology (ICCIT), 2012 15th International Conference on
Conference_Location :
Chittagong
Print_ISBN :
978-1-4673-4833-1
Type :
conf
DOI :
10.1109/ICCITechn.2012.6509701
Filename :
6509701
Link To Document :
بازگشت