Sixth European Conference on Speech Communication and Technology

In this paper, a theoretical framework is proposed for the introduction of the KNN pdf estimator in an HMMbased speech recognition system. The estimation of the state output probabilities with the KNN pdf estimator is shown to imply the introduction of a new parameter : the membership coefficient. To learn this coefficient with the BaumWelch algorithm, a maximum likelihood (ML) reestimation formula is derived. This new formula is tested and compared with the formula we had introduced previously [1]. Then, the edition/condensation techniques are introduced in the context of Markov models in an attempt to improve the appropriateness of the reference data set to the KNN HMM system. Two new algorithms are proposed for editing and condensing the reference set which present the advantage of being compatible with the KNN rule.
Full Paper (PDF) GnuZipped Postscript
Bibliographic reference. Lefévre, Fabrice / Montacié, Claude / Caraty, MarieJosé (1999): "A MLE algorithm for the kNN HMM system", In EUROSPEECH'99, 27332736.