DocumentCode :
1929761
Title :
Soft-decision vector quantization based on the Dempster/Shafer theory
Author :
Class, F. ; Kaltenmeier, A. ; Regel, P.
Author_Institution :
Daimler Benz AG Res. Inst., Ulm, Germany
fYear :
1991
fDate :
14-17 Apr 1991
Firstpage :
665
Abstract :
The authors describe an algorithm for soft-decision vector quantization (SVQ) implemented in the acoustic front-end of a large-vocabulary speech recognizer based on discrete density HMMs (hidden Markov models) of small phonetic units. In contrast to hard-decision vector quantization (HVQ), the proposed approach transforms a feature vector into a number of symbols associated with credibility values computed according to statistical models of distances and evidential reasoning. SVQ is related to semi-continuous density HMMs (SCHMMs). In contrast to SCHMM, which is based on multidimensional, class-specific distributions of feature vectors, SVQ is based on one-dimensional distributions of distances and is therefore much simpler. Credibilities and associated symbols form the inputs to both the HMM-training and the recognition modules of the system. SVQ improves recognition results remarkably
Keywords :
Markov processes; data compression; encoding; speech analysis and processing; speech recognition; Dempster/Shafer theory; HMM training; acoustic front-end; credibility values; discrete density HMM; distance distributions; feature vector; hidden Markov models; large-vocabulary speech recognizer; phonetic units; recognition modules; recognition results; semicontinuous density HMM; soft-decision vector quantization; speech coding; speech recognition; statistical models; Books; Hidden Markov models; Histograms; Multidimensional systems; Speech recognition; Vector quantization; Vocabulary;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on
Conference_Location :
Toronto, Ont.
ISSN :
1520-6149
Print_ISBN :
0-7803-0003-3
Type :
conf
DOI :
10.1109/ICASSP.1991.150427
Filename :
150427
Link To Document :
بازگشت