Paper Info
Title
Higher order cepstral moment normalization (HOCMN) for robust speech recognition
Abstract
Cepstral mean subtraction (CMS) and cepstral normalization (CN) have been popularly used to normalize the first and the second moments of cepstral coefficients, and proved to be very helpful for robust speech recognition (Furui, S. 1981; Viikki, O. and Laurila, K., 1998). A unified formulation for higher order cepstral moment normalization (HOCMN) is developed by extending the concept of CMS and CN to orders much higher than three. A whole family of normalization techniques for different orders is thus proposed. Preliminary experimental results based on Aurora 2.0 showed that the recognition accuracy can be significantly improved with this approach under all noisy conditions. For example, HOCMN(1,5,100) (normalization of the first, fifth and 100th order cepstral moments) is shown to offer an error rate reduction of 32.83% as compared to the conventional CN with a full-utterance processing interval, or an error rate reduction of 20.78% as compared to CN with a segmental processing interval.
Year
DOI
Venue
2004
10.1109/ICASSP.2004.1325956
ICASSP (1)
Keywords
DocType
Volume
speech recognition,cepstral normalization,higher order cepstral moment normalization,acoustic noise,cepstral analysis,cepstral mean subtraction,random noise,error rate reduction,robust speech recognition,error statistics,robustness,error rate,higher order,neural networks,noise reduction,mel frequency cepstral coefficient,probability density function,low frequency noise
Conference
1
ISSN
ISBN
Citations 
1520-6149
0-7803-8484-9
21
PageRank 
References 
Authors
1.08
2
2
Name
Order
Citations
PageRank
Chang-Wen Hsu1392.50
Lin-shan Lee21525182.03