DocumentCode :
302082
Title :
High-accuracy connected digit recognition for mobile applications
Author :
Gupta, Sunil K. ; Soong, Frank ; Haimi-Cohen, Raziel
Author_Institution :
Adv. Multi-Media Commun. Dept., AT&T Bell Labs., Middletown, NJ, USA
Volume :
1
fYear :
1996
fDate :
7-10 May 1996
Firstpage :
57
Abstract :
We present a connected digit recognition system with low storage and computational complexity which achieves good performance in car noise. Our system uses the TI-DIGITS database with additive car noise for training whole-word digit and background models. A digit accuracy of 96.1% is obtained on a 15-speaker database collected in a car using an open microphone with an average SNR of approximately 2 dB. There is a further error reduction of almost 35% if the top two candidate strings are considered using a traceback based N-best algorithm. The system can be implemented on a currently available fixed-point DSP chip. We show that significant performance improvements are obtained by using two-level cepstral mean subtraction (CMS), gender-dependent models and a decoding grammar constraining the possible lengths of digit strings
Keywords :
acoustic noise; automobiles; cepstral analysis; computational complexity; decoding; digital arithmetic; digital signal processing chips; land mobile radio; microphones; speech processing; speech recognition; TI-DIGITS database; additive car noise; average SNR; background models; connected digit recognition system; constrained length decoding; decoding grammar; digit accuracy; digit strings; error reduction; fixed-point DSP chip; gender dependent models; low computational complexity; low storage; mobile applications; open microphone; system performance; traceback based N-best algorithm; training; two-level cepstral mean subtraction; whole-word digit models; Additive noise; Background noise; Cepstral analysis; Collision mitigation; Computational complexity; Databases; Decoding; Digital signal processing chips; Microphones; Signal to noise ratio;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on
Conference_Location :
Atlanta, GA
ISSN :
1520-6149
Print_ISBN :
0-7803-3192-3
Type :
conf
DOI :
10.1109/ICASSP.1996.540289
Filename :
540289
Link To Document :
بازگشت