Sixth European Conference on Speech Communication and Technology
In the conventional hidden Markov model, the model parameters are reestimated by an iterative procedure known as the Baum-Welch method. This paper proposes an alternative procedure using fuzzy estimation, which is generalised from the fuzzy c-means and the Baum-Welch methods. An extension of this approach, which uses a garbage state to deal with outlier data is also proposed. Experiments using the TI46 speech data corpus show this approach can be applicable to speech and speaker recognition.
Full Paper (PDF) Gnu-Zipped Postscript
Bibliographic reference. Tran, Dat / Wagner, Michael (1999): "Hidden Markov models using fuzzy estimation", In EUROSPEECH'99, 2749-2752.