Hy Murveit

Genones: generalized mixture tying in continuous hidden Markov model-based speech recognizers

An algorithm is proposed that achieves a good tradeoff between modeling resolution and robustness by using a new, general scheme for tying of mixture components in continuous mixture-density hidden Markov model (HMM)-based speech recognizers. The sets of HMM states that share the same mixture components are determined automatically using agglomerative clustering techniques. Experimental results on ARPAs Wall Street Journal corpus show that this scheme reduces errors by 25% over typical tied-mixture systems. New fast algorithms for computing Gaussian likelihoods-the-most time-consuming aspect of continuous-density HMM systems-are also presented. These new algorithms-significantly reduce the number of Gaussian densities that are evaluated with little or no impact on speech recognition accuracy.

Large-vocabulary dictation using SRI's DECIPHER speech recognition system: progressive search techniques

The authors describe a technique called progressive search which is useful for developing and implementing speech recognition systems with high computational requirements. The scheme iteratively uses more and more complex recognition schemes, where each iteration constrains the speech space of the next. An algorithm called the forward-backward word-life algorithm is described. It can generate a word lattice in a progressive search that would be used as a language model embedded in a succeeding recognition pass to reduce computation requirements. It is shown that speed-ups of more than an order of magnitude are achievable with only minor costs in accuracy.<<ETX>>

Linguistic constraints in hidden Markov model based speech recognition

A speaker-independent, continuous-speech, large-vocabulary speech recognition system, DECIPHER, has been developed. It provides state-of-the-art performance on the DARPA standard speaker-independent resource management training and testing materials. The approach is to integrate speech and linguistic knowledge into the HMM (hidden Markov model) framework. Performance improvements arising from detailed phonological modeling and from the incorporation of cross-word coarticulatory constraints are described. It is concluded that speech and linguistic knowledge sources can be used to improve the performance of HMM-based speech recognition systems provided that care is taken to incorporate these knowledge sources appropriately.<<ETX>>

Genones: optimizing the degree of mixture tying in a large vocabulary hidden Markov model based speech recognizer

We propose a scheme that improves the robustness of continuous HMM systems that use mixture observation densities by sharing the same mixture components among different HMM states. The sets of HMM states that share the same mixture components are determined automatically using agglomerative clustering techniques. Experimental results on the Wall-Street Journal Corpus show that our new form of output distributions achieves a 25% reduction in error rate over typical tied-mixture systems.<<ETX>>

Spontaneous speech effects in large vocabulary speech recognition applications

We describe three analyses on the effects of spontaneous speech on continuous speech recognition performance. We have found that: (1) spontaneous speech effects significantly degrade recognition performance, (2) fluent spontaneous speech yields word accuracies equivalent to read speech, and (3) using spontaneous speech training data can significantly improve performance for recognizing spontaneous speech. We conclude that word accuracy can be improved by explicitly modeling spontaneous effects in the recognizer, and by using as much spontaneous speech training data as possible. Inclusion of read speech training data, even within the task domain, does not significantly improve performance.

Reduced channel dependence for speech recognition

Speech recognition systems tend to be sensitive to unimportant steady-state variation in speech spectra (i.e. those caused by varying the microphone or channel characteristics). There have been many attempts to solve this problem; however, these techniques are often computationally burdensome, especially for real-time implementation. Recently, Hermansy et al. [1] and Hirsch et al. [2] have suggested a simple technique that removes slow-moving linear channel variation with little adverse effect on speech recognition performance. In this paper we examine this technique, known as RASTA filtering, and evaluate its performance when applied to SRIs DECIPHER™ speech recognition system [3]. We show that RASTA filtering succeeds in reducing DECIPHER™s dependence on the channel.

Techniques to achieve an accurate real-time large-vocabulary speech recognition system

In addressing the problem of achieving high-accuracy real-time speech recognition systems, we focus on recognizing speech from ARPAs 20,000-word Wall Street Journal (WSJ) task, using current UNIX workstations. We have found that our standard approach---using a narrow beam width in a Viterbi search for simple discrete-density hidden Markov models (HMMs)---works in real time with only very low accuracy. Our most accurate algorithms recognize speech many times slower than real time. Our (yet unattained) goal is to recognize speech in real time at or near full accuracy.We describe the speed/accuracy trade-offs associated with several techniques used in a one-pass speech recognition framework:• Trade-offs associated with reducing the acoustic modeling resolution of the HMMs (e.g., output-distribution type, number of parameters, cross-word modeling)• Trade-offs associated with using lexicon trees, and techniques for implementing full and partial bigram grammars with those trees• Computation of Gaussian probabilities are the most time-consuming aspect of our highest accuracy system, and techniques allowing us to reduce the number of Gaussian probabilities computed with little or no impact on speech recognition accuracy.Our results show that tree-based modeling techniques used with appropriate acoustic modeling approaches achieve real-time performance on current UNIX workstations at about a 30% error rate for the WSJ task. The results also show that we can dramatically reduce the computational complexity of our more accurate but slower modeling alternatives so that they are near the speed necessary for real-time performance in a multipass search. Our near-future goal is to combine these two technologies so that real-time, high-accuracy large-vocabulary speech recognition can be achieved.

The decipher speech recognition system

A large-vocabulary, continuous-speech system, called DECIPHER, which is based on a hidden Markov model (HMM) approach and is designed to achieve high word accuracy in a speaker-independent mode is described. The results of a series of experiments that test acoustic and phonological adaptation of the DECIPHER system to the pronunciations of a single speaker in a speaker-dependent task are presented. Estimating the probabilities of alternative pronunciations and speaker-dependent phonology are discussed.<<ETX>>

Integrating speech and natural-language processing

SRI has developed a new architecture for integrating speech and natural-language processing that applies linguistic constraints during recognition by incrementally expanding the state-transition network embodied in a unification grammar. We compare this dynamic-grammar-network (DGN) approach to its principal alternative, word-lattice parsing, presenting preliminary experimental results that suggest the DGN approach requires much less computation time than word-lattice parsing, while maintaining a very tractable recognition search space.

SRI's DECIPHER system

SRI has developed a speaker-independent continuous speech, large vocabulary speech recognition system, DECIPHER, that provides state-of-the-art performance on the DARPA standard speaker-independent resource management training and testing materials. SRIs approach is to integrate speech and linguistic knowledge into the HMM framework. This paper describes performance improvements arising from detailed phonological modeling and from the incorporation of cross-word coarticulatory constraints.