Abstract | ||
---|---|---|
An extension of the multi-band model including inter-band control of time asynchrony is described. It is based on the framework of Markov random fields. The law of the speech process is given by a parametric Gibbs distribution and a maximum likelihood parameter estimation algorithm is developed. This random field model is applied to isolated word recognition. It is shown that similar performances are obtained with the new model and with standard HMM techniques in the mono-band case. In the multi-band case, it is shown that the recognition rate decreases when the number of bands is increased but that modeling inter-band synchrony limits the performance decrease |
Year | DOI | Venue |
---|---|---|
2000 | 10.1109/ICASSP.2000.862013 | ICASSP |
Keywords | Field | DocType |
parametric gibbs distribution,maximum likelihood parameter estimation algorithm,inter-band synchrony,speech recognition,random processes,time asynchrony,maximum likelihood estimation,inter-band control,mono-band case,markov random field based multi-band model,isolated word recognition,speech process,modeling,markov processes,recognition rate,performance,gibbs distribution,random field,lattices,speech processing,parameter estimation,word recognition,hidden markov models,maximum likelihood | Random field,Maximum-entropy Markov model,Markov property,Pattern recognition,Markov model,Markov random field,Computer science,Markov chain,Artificial intelligence,Variable-order Markov model,Hidden Markov model | Conference |
Volume | ISSN | ISBN |
3 | 1520-6149 | 0-7803-6293-4 |
Citations | PageRank | References |
2 | 0.39 | 5 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
guillaume gravier | 1 | 1413 | 127.38 |
Marc Sigelle | 2 | 316 | 34.12 |
Gérard Chollet | 3 | 725 | 129.74 |