On the use of lattices for the automatic generation of pronunciations

In this paper, we explore the use of lattices to generate pronunciations for speech recognition based on the observation of a few (say one or two) speech utterances of a word. Various search strategies are investigated in combination with schemes where single or multiple pronunciations are generated...

Full description

Saved in:
Bibliographic Details
Published in2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03) Vol. 1; p. I
Main Authors Deligne, S., Mangu, L.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2003
Subjects
Cepstral analysis
Context modeling
Decision trees
Gaussian distribution
Gaussian processes
Lattices
Merging
Speech recognition
Viterbi algorithm
Vocabulary
Online AccessGet full text

Cover

More Information
Summary:In this paper, we explore the use of lattices to generate pronunciations for speech recognition based on the observation of a few (say one or two) speech utterances of a word. Various search strategies are investigated in combination with schemes where single or multiple pronunciations are generated for each speech utterance. In our experiments, a strategy that combines merging time-overlapping links in a context-dependent subphone lattice and generating multiple pronunciations provides the best recognition accuracy. This results in average relative gains of 30% over the generation of single pronunciations using a Viterbi search.
ISBN:9780780376632
0780376633
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2003.1198752