A study on speaker adaptation of continuous density HMM parameters

Lee, C.-H.; Lin, Che-Kuang; Juang, Biing‐Hwang

doi:10.1109/icassp.1990.115559

Cited by 23 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is of course a general formulation that holds for any model transformation. The model transformation parameters are first derived under a MAP criterion according to (7). The model parameters are then transformed and reestimated according to (8), using the priors of the transformed models (that point will be further discussed in Section IV).…”

Section: )mentioning

confidence: 99%

See 1 more Smart Citation

Joint maximum a posteriori adaptation of transformation and HMM parameters

Siohan¹,

Chesta

Lee³

2001

IEEE Trans. Speech Audio Process.

View full text Add to dashboard Cite

Model adaptation techniques are an efficient way to reduce the mismatch that typically occurs between the training and test condition of any speech recognizer. Adaptation techniques can usually be divided into two families of approaches. On one hand, direct model adaptation attempts to directly reestimate the model parameters, for example using MAP adaptation. Since direct adaptation only reestimates model parameters of the corresponding units appearing in the adaptation data, a large amount of such data is needed to observe any significant improvement in performance. However, nice asymptotic properties are usually observed, meaning that the performance improves as the amount of adaptation data increases. On the other hand, indirect model adaptation applies a general transformation on some clusters of model parameters. Because each individual model is transformed, the approach is quite effective when a small amount of adaptation data is available. However, as the amount of adaptation data increases, the performance improvement quickly saturates. In this paper, we propose to jointly estimate model parameters and transformation parameters using a single estimation criterion based on Bayesian statistics. We show that by providing a prior distribution for the model parameters and the transformation parameters, it is possible to jointly estimate these two sets of parameters using maximum a posteriori estimation (MAP). Experimental evaluation on nonnative speaker and channel adaptation illustrates the effectiveness of the proposed approach.

show abstract

Section: )mentioning

confidence: 99%

“…Amongst direct model adaptation techniques, Bayesian parameter learning is the most representative approach [7], [8].…”

Section: Introductionmentioning

confidence: 99%

Joint maximum a posteriori adaptation of transformation and HMM parameters

Siohan¹,

Chesta

Lee³

2001

IEEE Trans. Speech Audio Process.

View full text Add to dashboard Cite

show abstract

“…However, these schemes make assumptions about the form of the acoustic environment. Techniques that only update distributions for which observations occur in the adaptation data, such as those using maximum a posteriori (MAP) estimation (Lee, Lin & Juang, 1990;Gauvain & Lee, 1994), require a relatively large amount of adaptation data to be effective. An alternative approach is to estimate a set of transformations that can be applied to the model parameters.…”

Section: Introductionmentioning

confidence: 99%

Mean and variance adaptation within the MLLR framework

Gales

Woodland

1996

Computer Speech & Language

330

175

View full text Add to dashboard Cite

“…Thus, if a mixture component is not observed in the adaptation data, then it cannot be adapted. If we wish to update all the model parameters, then we require a fairly large amount of training data (Gauvain & Lee 1994;Lee et al 1990). Therefore, this approach is less commonly used when compared to the ML approach discussed below.…”

Section: Speaker Adaptation Approach To Normalizationmentioning

confidence: 98%

Studies on inter-speaker variability in speech and its application in automatic speech recognition

Umesh

2011

Sadhana

View full text Add to dashboard Cite

In this paper, we give an overview of the problem of inter-speaker variability and its study in many diverse areas of speech signal processing. We first give an overview of vowel-normalization studies that minimize variations in the acoustic representation of vowel realizations by different speakers. We then describe the universal-warping approach to speaker normalization which unifies many of the vowel normalization approaches and also shows the relation between speech production, perception and auditory processing. We then address the problem of inter-speaker variability in automatic speech recognition (ASR) and describe techniques that are used to reduce these effects and thereby improve the performance of speaker-independent ASR systems.

show abstract

A study on speaker adaptation of continuous density HMM parameters

Cited by 23 publications

References 8 publications

Joint maximum a posteriori adaptation of transformation and HMM parameters

Joint maximum a posteriori adaptation of transformation and HMM parameters

Mean and variance adaptation within the MLLR framework

Studies on inter-speaker variability in speech and its application in automatic speech recognition

Contact Info

Product

Resources

About