From HMM's to segment models: a unified view of stochastic modeling for speech recognition

Ostendorf, Mari; Digalakis, Vassilios; Kimball, Owen

doi:10.1109/89.536930

Cited by 467 publications

(308 citation statements)

References 74 publications

Supporting

Mentioning

305

Contrasting

Unclassified

Order By: Relevance

“…A possible alternative to HMM are the Stochastic Segmental Models. The more sophisticated segmental techniques fit parametric curves to the feature trajectories of the phonemes [24]. There is, however, a much simpler methodology [13] that applies non-uniform smoothing and sampling in order to parametrize any phoneme with the same number of features, independent of its length.…”

Section: Hidden Markov and Segmental Phoneme Modelingmentioning

confidence: 99%

Application of Kernel-Based Feature Space Transformations and Learning Methods to Phoneme Classification

Kocsor

Tóth

2004

Applied Intelligence

View full text Add to dashboard Cite

Abstract. This paper examines the applicability of some learning techniques to the classification of phonemes. The methods tested were artificial neural nets (ANN), support vector machines (SVM) and Gaussian mixture modeling (GMM). We compare these methods with a traditional hidden Markov phoneme model (HMM), working with the linear prediction-based cepstral coefficient features (LPCC). We also tried to combine the learners with linear/nonlinear and unsupervised/supervised feature space transformation methods such as principal component analysis (PCA), independent component analysis (ICA), linear discriminant analysis (LDA), springy discriminant analysis (SDA) and their nonlinear kernel-based counterparts. We found that the discriminative learners can attain the efficiency of HMM, and that after the transformations they can retain the same performance in spite of the severe dimension reduction. The kernel-based transformations brought only marginal improvements compared to their linear counterparts.

show abstract

Section: Hidden Markov and Segmental Phoneme Modelingmentioning

confidence: 99%

Application of Kernel-Based Feature Space Transformations and Learning Methods to Phoneme Classification

Kocsor

Tóth

2004

Applied Intelligence

View full text Add to dashboard Cite

show abstract

“…Because MPCs exploit the notion that trajectories are continuous, they also bear some resemblance to so-called segmental HMMs (Ostendorf, Digalakis, & Kimball, 1996). MPCs nevertheless differ from segmental HMMs in two important respects: (i) the treatment of arc length-particularly, the estimation of a metric g i (x) for each hidden state of the Markov process, and (ii) the natural parameterization of a segmentation model Pr[s | x], as opposed to a synthesis model, Pr [x | s], that is even more complicated than the one in ordinary HMMs.…”

Section: Relation To Hidden Markov Models and Previous Workmentioning

confidence: 99%

Untitled

Saul

Rahim

2000

Machine Learning

View full text Add to dashboard Cite

Abstract.We study the classification problem that arises when two variables-one continuous (x), one discrete (s)-evolve jointly in time. We suppose that the vector x traces out a smooth multidimensional curve, to each point of which the variable s attaches a discrete label. The trace of s thus partitions the curve into different segments whose boundaries occur where s changes value. We consider how to learn the mapping between the trace of x and the trace of s from examples of segmented curves. Our approach is to model the conditional random process that generates segments of constant s along the curve of x. We suppose that the variable s evolves stochastically as a function of the arc length traversed by x. Since arc length does not depend on the rate at which a curve is traversed, this gives rise to a family of Markov processes whose predictions are invariant to nonlinear warpings (or reparameterizations) of time. We show how to estimate the parameters of these models-known as Markov processes on curves (MPCs)-from labeled and unlabeled data. We then apply these models to two problems in automatic speech recognition, where x are acoustic feature trajectories and s are phonetic alignments.

show abstract

“…To release from the limitations of HMM, some alternative models have been proposed. One of the models is the segment model (SM) [1].…”

Section: Introductionmentioning

confidence: 99%

Landmark-Guided Segmental Speech Decoding for Continuous Mandarin Speech Recognition

2016

J Inf Process Syst

View full text Add to dashboard Cite

In this paper, we propose a framework that attempts to incorporate landmarks into a segment-based Mandarin speech recognition system. In this method, landmarks provide boundary information and phonetic class information, and the information is used to direct the decoding process. To prove the validity of this method, two kinds of landmarks that can be reliably detected are used to direct the decoding process of a segment model (SM) based Mandarin LVCSR (large vocabulary continuous speech recognition) system. The results of our experiment show that about 30% decoding time can be saved without an obvious decrease in recognition accuracy. Thus, the potential of our method is demonstrated.

show abstract

From HMM's to segment models: a unified view of stochastic modeling for speech recognition

Cited by 467 publications

References 74 publications

Application of Kernel-Based Feature Space Transformations and Learning Methods to Phoneme Classification

Application of Kernel-Based Feature Space Transformations and Learning Methods to Phoneme Classification

Untitled

Landmark-Guided Segmental Speech Decoding for Continuous Mandarin Speech Recognition

Contact Info

Product

Resources

About