Acoustic factorisation

Gales, Mjf

doi:10.1109/asru.2001.1034593

Cited by 33 publications

(39 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work will also use speaker-dependent Jacobians, but in a fully model-based framework. For model-based approaches, acoustic factorisation [9] (e.g. structured transform [10]) has been proposed.…”

Section: Introductionmentioning

confidence: 99%

Speaker and noise factorisation on the AURORA4 task

Wang

Gales

2011

2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

For many realistic scenarios, there are multiple factors that affect the clean speech signal. In this work approaches to handling two such factors, speaker and background noise differences, simultaneously are described. A new adaptation scheme is proposed. Here the acoustic models are first adapted to the target speaker via an MLLR transform. This is followed by adaptation to the target noise environment via model-based vector Taylor series (VTS) compensation. These speaker and noise transforms are jointly estimated, using maximum likelihood. Experiments on the AURORA4 task demonstrate that this adaptation scheme provides improved performance over VTS-based noise adaptation. In addition, this framework enables the speech and noise to be factorised, allowing the speaker transform estimated in one noise condition to be successfully used in a different noise condition.

show abstract

Section: Introductionmentioning

confidence: 99%

Speaker and noise factorisation on the AURORA4 task

Wang

Gales

2011

2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…However, for found data, there may be multiple acoustic factors affecting the speech signal. This motivates the use of multiple forms of transformations, denoted here as STs, to represent complex nonspeech variabilities in an adaptive training framework [16], [24]. In this paper, one particular form of ST is investigated which is appropriate for state-of-the-art speech recognition systems.…”

Section: Structured Transformsmentioning

confidence: 99%

“…This form of I-smoothing prior is similar to the smoothing function (17). The log prior may be expressed as (24) where is the general form defined in (15) and…”

Section: I-smoothing Prior Distributionmentioning

confidence: 99%

Discriminative cluster adaptive training

Gales²

2006

IEEE Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

Abstract-Multiple-cluster schemes, such as cluster adaptive training (CAT) or eigenvoice systems, are a popular approach for rapid speaker and environment adaptation. Interpolation weights are used to transform a multiple-cluster, canonical, model to a standard hidden Markov model (HMM) set representative of an individual speaker or acoustic environment. Maximum likelihood training for CAT has previously been investigated. However, in state-of-the-art large vocabulary continuous speech recognition systems, discriminative training is commonly employed. This paper investigates applying discriminative training to multiple-cluster systems. In particular, minimum phone error (MPE) update formulae for CAT systems are derived. In order to use MPE in this case, modifications to the standard MPE smoothing function and the prior distribution associated with MPE training are required. A more complex adaptive training scheme combining both interpolation weights and linear transforms, a structured transform (ST), is also discussed within the MPE training framework. Discriminatively trained CAT and ST systems were evaluated on a state-of-the-art conversational telephone speech task. These multiple-cluster systems were found to outperform both standard and adaptively trained systems.Index Terms-Cluster adaptive training (CAT), discriminative training, eigenvoices, minimum phone error (MPE), multiple-cluster HMM.

show abstract

“…Using a Gaussian Mixture Model (GMM) of G Gaussians, the likelihood of the coefficients φ given tag T ag t is calculated as in Equation 4.…”

Section: Using Contextual Information In Eigenspace Mllrmentioning

confidence: 99%

“…Acoustic model factorisation techniques in ASR [4,5,6,7], which separate the many factors in an audio signal, are gaining relevance in dealing with diverse acoustic conditions, as they did before in speaker identification tasks [8,9]. Early techniques for providing factorisation in ASR were Cluster Adaptive Training (CAT) [10] and eigenspace MLLR [11,12].…”

Section: Introductionmentioning

confidence: 99%

Using contextual information in joint factor eigenspace MLLR for speech recognition in diverse scenarios

Saz

Hain

2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper presents a new approach for rapid adaptation in the presence of highly diverse scenarios that takes advantage of information describing the input signals. We introduce a new method for joint factorisation of the background and the speaker in an eigenspace MLLR framework: Joint Factor Eigenspace MLLR (JFEMLLR). We further propose to use contextual information describing the speaker and background, such as tags or more complex metadata, to provide an immediate estimation of the best MLLR transformation for the utterance. This provides instant adaptation, since it does not require any transcription from a previous decoding stage. Evaluation in a highly diverse Automatic Speech Recognition (ASR) task, a modified version of WSJCAM0, yields an improvement of 26.9% over the baseline, which is an extra 1.2% reduction over two-pass MLLR adaptation.

show abstract

Acoustic factorisation

Cited by 33 publications

References 3 publications

Speaker and noise factorisation on the AURORA4 task

Speaker and noise factorisation on the AURORA4 task

Discriminative cluster adaptive training

Using contextual information in joint factor eigenspace MLLR for speech recognition in diverse scenarios

Contact Info

Product

Resources

About