DocumentCode :
3326295
Title :
Spectral entropy based feature for robust ASR
Author :
Misra, Hemant ; Ikbal, Shajith ; Bourlard, Hervé ; Hermansky, Hynek
Author_Institution :
Dalle Molle Inst. for Perceptual Artificial Intelligence, Martigny, Switzerland
Volume :
1
fYear :
2004
fDate :
17-21 May 2004
Abstract :
In general, entropy gives us a measure of the number of bits required to represent some information. When applied to the probability mass function (PMF), entropy can also be used to measure the "peakiness" of a distribution. We propose using the entropy of a short time Fourier transform spectrum, normalised as PMF, as an additional feature for automatic speech recognition (ASR). It is indeed expected that a peaky spectrum, representation of clear formant structure in the case of voiced sounds, will have low entropy, while a flatter spectrum, corresponding to nonspeech or noisy regions, will have higher entropy. Extending this reasoning further, we introduce the idea of a multiband/multiresolution entropy feature where we divide the spectrum into equal size subbands and compute entropy in each subband. The results show that multiband entropy features used in conjunction with normal cepstral features improve the performance of an ASR system.
Keywords :
Fourier transforms; entropy; spectral analysis; speech recognition; statistical distributions; automatic speech recognition; cepstral features; flat spectrum; formant structure; multiband entropy feature; multiresolution entropy feature; peaky spectrum; probability mass function; robust ASR; short time Fourier transform spectrum; spectral entropy features; voiced sounds; Acoustic noise; Artificial intelligence; Automatic speech recognition; Cepstral analysis; Energy capture; Entropy; Fourier transforms; Information theory; Mel frequency cepstral coefficient; Robustness;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on
ISSN :
1520-6149
Print_ISBN :
0-7803-8484-9
Type :
conf
DOI :
10.1109/ICASSP.2004.1325955
Filename :
1325955
Link To Document :
بازگشت