Using Prosodic Features in Language Models for Meetings

Using Prosodic Features in Language Models for Meetings

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Songfang Huang , Steve Renals

Citations:

4 - 0 self

Versions

Version 0

Version History

Metadata Version 0

Datum	Value	Source
TITLE	Using Prosodic Features in Language Models for Meetings	SVM HeaderParse 0.2
AUTHOR NAME	Songfang Huang	SVM HeaderParse 0.2
AUTHOR AFFIL	The Centre for Speech Technology Research; University of Edinburgh	SVM HeaderParse 0.2
AUTHOR ADDR	Edinburgh, EH8 9LW, UK	SVM HeaderParse 0.2
AUTHOR NAME	Steve Renals	SVM HeaderParse 0.2
AUTHOR AFFIL	The Centre for Speech Technology Research; University of Edinburgh	SVM HeaderParse 0.2
AUTHOR ADDR	Edinburgh, EH8 9LW, UK	SVM HeaderParse 0.2
ABSTRACT	Abstract. Prosody has been actively studied as an important knowledge source for speech recognition and understanding. In this paper, we are concerned with the question of exploiting prosody for language models to aid automatic speech recognition in the context of meetings. Using an automatic syllable detection algorithm, the syllable-based prosodic features are extracted to form the prosodic representation for each word. Two modeling approaches are then investigated. One is based on a factored language model, which directly uses the prosodic representation and treats it as a ‘word’. Instead of direct association, the second approach provides a richer probabilistic structure within a hierarchical Bayesian framework by introducing an intermediate latent variable to represent similar prosodic patterns shared by groups of words. Fourfold cross-validation experiments on the ICSI Meeting Corpus show that exploiting prosody for language modeling can significantly reduce the perplexity, and also have marginal reductions in word error rate. 1	SVM HeaderParse 0.2
CITATIONS	22 found	ParsCit 1.0