Author: SRI International

  • Acoustic Clustering and Adaptation for Robust Speech Recognition

    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.

  • Speech: A Privileged Modality

    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. 

  • Modeling Linguistic Segment and Turn Boundaries for N-best Rescoring of Spontaneous Speech

    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.

  • A Lognormal Tied Mixture Model of Pitch for Prosody-Based Speaker Recognition

    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.

  • Explicit Word Error Minimization in N-best List Rescoring

    We show that the standard hypothesis scoring paradigm used in maximum-likelihood-based speech recognition systems is not optimal with regard to minimizing the word error rate, the commonly used performance metric in speech recognition.

  • A Study of Multilingual Speech Recognition

    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.

  • Diagrammatic Methods for Deriving and Relating Temporal Neural Network Algorithms

    We present an alternative approach based on a set of simple block diagram manipulation rules. The approach provides a common framework to derive popular algorithms including backpropagation and backpropagation-through-time, without a single chain rule expansion.

  • Structure and Performance of a Dependency Language Model

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

  • Multimodal Interfaces for Internet

    In this paper, we present a Java-enabled application with a multimodal (pen and voice) interface over the web. Our implementation approach was to add Java to the set of languages accepted by the Open Agent Architecture (OAA), a framework for rapidly prototyping complex applications, and particularly suited to those with multimodal interfaces.

  • Using Differential Constraints to Reconstruct Complex Surfaces from Stereo

    Stereo reconstruction algorithms often fail to properly deal with complex surfaces, because there is not enough image information. We propose to guide the reconstruction process using a priori information about the differential geometry of the object surfaces.

  • Handset-Dependent Background Models for Robust Text-Independent Speaker Recognition

    This paper studies the effects of handset distortion on telephone-based speaker recognition performance. Results on the 1996 NIST Speaker Recognition Evaluation corpus show that using handset-matched background models reduces false acceptances (at a 10% miss rate) by more than 60% over previously reported (handset-independent) approaches.

  • Neural-Network Based Measures of Confidence for Word Recognition

    This paper proposes a probabilstic framework to define and evaluate confidence measures for word recognition. We describe a novel method to combine different knowledge sources and estimate the confidence in a word hypothesis, via a neural network.