4th International Conference on Spoken Language Processing

Philadelphia, PA, USA
October 3-6, 1996

Likelihood Ratio Decoding and Confidence Measures for Continuous Speech Recognition

Eduardo Lleida-Solano (1), Richard C. Rose (2)

(1) University of Zaragoza, Spain; (2) AT&T Research, Murray Hill, NJ, USA

Automatic speech recognition (ASR) systems are being integrated into a wider variety of tasks involving human{machine interaction. In evaluating these systems, however, it has become clear that more accurate means must be developed for detecting when portions of the decoded recognition hypotheses are either incorrect or represent out{of{vocabulary utterances. This paper describes the use of confidence measures based on likelihood ratio based optimization procedures for decoding and rescoring word hypotheses in an HMM based speech recognizer. These techniques are applied to spontaneous utterances obtained from a \movie locator" based dialog task.

Full Paper

Bibliographic reference.  Lleida-Solano, Eduardo / Rose, Richard C. (1996): "Likelihood ratio decoding and confidence measures for continuous speech recognition", In ICSLP-1996, 478-481.