On modeling duration in context in speech recognition
- 13 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 421-424
- https://doi.org/10.1109/icassp.1989.266455
Abstract
A clustering algorithm is introduced that allows clustering of HMM (hidden Markov models) models directly. This clustering algorithm determines the appropriate duration profile for a recognition unit. High-performance speaker-independent digit recognition on a studio-quality connected-digit database is demonstrated using this algorithm.Keywords
This publication has 5 references indexed in Scilit:
- Low rate speech coding using contour quantizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- A database for speaker-independent digit recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- The LPC trace as an HMM development toolThe Journal of the Acoustical Society of America, 1988
- A model-based connected-digit recognition system using either hidden Markov models or templatesComputer Speech & Language, 1986
- Interactive clustering techniques for selecting speaker-independent reference templates for isolated word recognitionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1979