Ginisupport vector machines for segmental minimum Bayes risk decoding of continuous speech

Veera Venkataramani, Shantanu Chakrabartty, William Byrne

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

We describe the use of support vector machines (SVMs) for continuous speech recognition by incorporating them in segmental minimum Bayes risk decoding. Lattice cutting is used to convert the Automatic Speech Recognition search space into sequences of smaller recognition problems. SVMs are then trained as discriminative models over each of these problems and used in a rescoring framework. We pose the estimation of a posterior distribution over hypotheses in these regions of acoustic confusion as a logistic regression problem. We also show that GiniSVMs can be used as an approximation technique to estimate the parameters of the logistic regression problem. On a small vocabulary recognition task we show that the use of GiniSVMs can improve the performance of a well trained hidden Markov model system trained under the Maximum Mutual Information criterion. We also find that it is possible to derive reliable confidence scores over the GiniSVM hypotheses and that these can be used to good effect in hypothesis combination. We discuss the problems that we expect to encounter in extending this approach to large vocabulary continuous speech recognition and describe initial investigation of constrained estimation techniques to derive feature spaces for SVMs.

Original languageEnglish
Pages (from-to)423-442
Number of pages20
JournalComputer Speech and Language
Volume21
Issue number3
DOIs
StatePublished - Jul 2007

Fingerprint

Dive into the research topics of 'Ginisupport vector machines for segmental minimum Bayes risk decoding of continuous speech'. Together they form a unique fingerprint.

Cite this