Revisiting semi-continuous hidden Markov models

K. Riedhammer, T. Bocklet, A. Ghoshal, D. Povey

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the past decade, semi-continuous hidden Markov models (SCHMMs) have not attracted much attention in the speech recognition community. Growing amounts of training data and increasing sophistication of model estimation led to the impression that continuous HMMs are the best choice of acoustic model. However, recent work on recognition of under-resourced languages faces the same old problem of estimating a large number of parameters from limited amounts of transcribed speech. This has led to a renewed interest in methods of reducing the number of parameters while maintaining or extending the modeling capabilities of continuous models. In this work, we compare classic and multiple-codebook semi-continuous models using diagonal and full covariance matrices with continuous HMMs and subspace Gaussian mixture models. Experiments on the RM and WSJ corpora show that while a classical semicontinuous system does not perform as well as a continuous one, multiple-codebook semi-continuous systems can perform better, particular when using full-covariance Gaussians.
Original languageEnglish
Title of host publicationAcoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages4721-4724
Number of pages4
DOIs
Publication statusPublished - 2012

Keywords

  • Gaussian processes
  • covariance matrices
  • hidden Markov models
  • parameter estimation
  • speech recognition
  • RM corpora
  • SCHMM
  • WSJ corpora
  • acoustic model
  • classical semicontinuous system
  • diagonal covariance matrices
  • full covariance matrices
  • full-covariance Gaussians
  • multiple-codebook semicontinuous models
  • semicontinuous hidden Markov model
  • speech recognition community
  • training data
  • under-resourced languages
  • Computational modeling
  • Data models
  • Hidden Markov models
  • Smoothing methods
  • Speech recognition
  • acoustic modeling
  • automatic speech recognition

Fingerprint Dive into the research topics of 'Revisiting semi-continuous hidden Markov models'. Together they form a unique fingerprint.

Cite this