Study of Associative Cepstral Statistics Normalization Techniques for Robust Speech Recognition in Additive Noise Environments

Wen-Hsiang Tu ； Jeih-Weih Hung

《中文計算語言學期刊》 14卷1期 (2009/03) Pp. 105-131

https://doi.org/10.30019/IJCLCLP.200903.0005

並列摘要

Feature statistics normalization techniques have been shown to be very successful in improving the noise robustness of a speech recognition system. In this paper, we propose an associative scheme in order to obtain a more accurate estimate of the statistical information in these techniques. By properly integrating codebook and utterance knowledge, the resulting associative cepstral mean subtraction (A-CMS), associative cepstral mean and variance normalization (A-CMVN), and associative histogram equalization (A-HEQ) behave significantly better than the conventional utterance-based and codebook-based versions in additive noise environments. For the Aurora-2 clean-condition training task, the new proposed associative histogram equalization (A-HEQ) provides an average recognition accuracy of 90.69%, which is better than utterance-based HEQ (87.67%) and codebook-based HEQ (86.00%).

並列關鍵字

Speech Recognition ； Noise-Robust Feature ； Codebook

參考文獻

Acero, A.(1990).Acoustical and environmental robustness in automatic speech recognition.Department of Electrical and Computer Engineering, Carnegie Mellon University.

Acero, A.,Deng, L.,Kristjansson, T.,Zhang, J.(2000).HMM adaptation using vector Taylor series for noisy speech recognition.Proceeding of 2000 International Conference on Spoken Language Processing (ICSLP 2000).(Proceeding of 2000 International Conference on Spoken Language Processing (ICSLP 2000)).

Atal, B.S.(1974).Effectiveness of linear prediction characteristics of the speech wave for automatic speaker identification and verification.Journal of the Acoustic Society of America.55,1304-1312.

Beattie, V. L.,Young, S. J.(1992).Hidden Markov model state-based cepstral noise compensation.Proceeding of International Conference on Spoken Language Processing (ICSLP 1992).(Proceeding of International Conference on Spoken Language Processing (ICSLP 1992)).

Berstein, A. D.,Shallom, I. D.(1991).An hypothesized Wiener filtering approach to noisy speech recognition.(Proceeding of 1991 International Conference on Acoustics, Speech and Signal Processing (ICASSP 1991)).