Author: SRI International

This paper presents a new training procedure for speaker verification systems. Results are presented from the 1997 NIST Speaker Recognition Evaluation corpus indicating that the VCF performance can be improved with this procedure.

We describe a new approach for statistical modeling and detection of discourse structure for natural conversational speech. Our model is based on 42 `Dialog Acts’ (DAs), (question, answer, backchannel, agreement, disagreement, apology, etc).

We present a maximum entropy language model that incorporates both syntax and semantics via a dependency grammar.

We have developed a disfluency detection method using decision tree classifiers that use only local and automatically extracted prosodic features. Because the model doesn’t rely on lexical information, it is widely applicable even when word recognition is unreliable.

We describe an algorithm based on acoustic clustering and acoustic adaptation to significantly improve speech recognition performance. The method is particularly useful when speech from multiple speakers is to be recognized and the boundary between speakers is not known.

In this article, we use our interaction model to demonstrate that during multimodal fusion, speech should be a privileged modality, driving the interpretation of a query, and that in certain cases, speech has even more power to override and modify the combination of other modalities than previously believed. 

We present an N-best rescoring algorithm that removes the effect of segmentation mismatch. Furthermore, we show that explicit language modeling of hidden linguistic segment boundaries is improved by including turn-boundary events in the model.

This paper describes our work in developing multilingual (Swedish and English) speech recognition systems in the ATIS domain. The acoustic component of the multilingual systems is realized through sharing Gaussian codebooks across Swedish and English allophones.

In this work, we develop a statistical model of pitch that allows unbiased estimation of pitch statistics from pitch tracks which are subject to doubling and/or halving.