On including temporal constraints in Viterbi alignment for speech recognition in noise

Página do item simplificado
dc.contributor.authorYoma N.B.
dc.contributor.authorMcInnes F.R.
dc.contributor.authorJack M.A.
dc.contributor.authorStump S.D.
dc.contributor.authorLing L.L.
dc.date.accessioned2024-03-13T01:46:57Z
dc.date.available2024-03-13T01:46:57Z
dc.date.issued2001
dc.description.abstractThis paper addresses the problem of temporal constraints in the Viterbi algorithm in speaker-dependent and independent tasks. The results here presented suggest that in a speaker-dependent task the introduction of temporal constraints can lead to a high improvement with additive or convolutional noise, the statistical modeling of state durations is not relevant if the max and min state duration restrictions are imposed, and truncated probability densities give better results than a metric previously proposed. Finally, word position dependent and independent temporal restrictions are compared in connected word speech recognition experiments and it is shown that the former leads to better results with the same computational load. However, duration model effect could be much less significant when the acoustic model is optimized and when the training and testing conditions are matched.
dc.description.firstpage179
dc.description.issuenumber2
dc.description.lastpage182
dc.description.volume9
dc.identifier.doi10.1109/89.902285
dc.identifier.issn1063-6676
dc.identifier.urihttps://dspace.mackenzie.br/handle/10899/38036
dc.relation.ispartofIEEE Transactions on Speech and Audio Processing
dc.rightsAcesso Restrito
dc.titleOn including temporal constraints in Viterbi alignment for speech recognition in noise
dc.typeArtigo
local.scopus.citations17
local.scopus.eid2-s2.0-0035249864
local.scopus.subjectHidden Markov models (HMM)
local.scopus.updated2024-05-01
local.scopus.urlhttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=0035249864&origin=inward
Arquivos
Coleções
Artigos de periódico