8th International Conference on Spoken Language Processing

Jeju Island, Korea
October 4-8, 2004

Acoustic Phonetic Modeling using Local Codebook Features

Frank Diehl, Asuncion Moreno

Universitat Politecnica de Catalunya, Spain

In this article we present an alternative method for defining the question set used for the induction of acoustic phonetic decision trees. The method is data driven and employs local similarities between the probability density functions of hidden Markov models. The method is shown to work at least as well as the standard method using question sets devised by human experts.

Full Paper

Bibliographic reference.  Diehl, Frank / Moreno, Asuncion (2004): "Acoustic phonetic modeling using local codebook features", In INTERSPEECH-2004, 709-712.