EUROSPEECH 2001 Scandinavia
7th European Conference on Speech Communication and Technology

Aalborg, Denmark
September 3-7, 2001


Evaluation of a Generalized Dynamic Cepstrum in Distant Speech Recognition

Hiroshi Matsumoto, Akihiko Shimizu, Kazumasa Yamamoto

Shinshu University, Japan

This paper examines the effectiveness of a generalized dynamic cepstrum in distant speech recognition. The generalized dynamic cepstrum (DyMFGC) is based upon the forward masking on the generalized logarithmic spectrum instead of the log-spectrum, which intends to make it robust to additive noise as well as convolutional noise. Digit recognition tests were carried out in a relatively quiet and small sized office environment. Under white noise environments, the DyMFGC outperforms the dynamic cepstrum on the logarithmic spectrum and the MFCC with cepstral mean normalization. It also maintains the word accuracy of 90% to 95% within a 1m distance from a source. In speech babble noise environments, the performance of the DyMFGC is approximately the same as that of the dynamic cepstrum on the logarithmic amplitude scale.

Full Paper

Bibliographic reference.  Matsumoto, Hiroshi / Shimizu, Akihiko / Yamamoto, Kazumasa (2001): "Evaluation of a generalized dynamic cepstrum in distant speech recognition", In EUROSPEECH-2001, 881-884.