Very fast adaptation with a compact context-dependent eigenvoice model

Kühn, Roland; Perronnin, Florent; Nguyen, Patrick; Junqua, Jean‐Claude; Rigazio, Luca

doi:10.1109/icassp.2001.940845

Cited by 15 publications

(17 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We call this a Subspace GMM (SGMM). This method has similarities to Joint Factor Analysis as used in speaker recognition (Kenny et al, 2008) and to Eigenvoices (Kuhn et al, 2001) and Cluster Adaptive Training (CAT) (Gales, 2001) as proposed for speech recognition.…”

Section: Introductionmentioning

confidence: 99%

The subspace Gaussian mixture model—A structured model for speech recognition

Povey

Burget

Agarwal

et al. 2011

Computer Speech & Language

257

195

View full text Add to dashboard Cite

We describe a new approach to speech recognition, in which all Hidden Markov Model (HMM) states share the same Gaussian Mixture Model (GMM) structure with the same number of Gaussians in each state. The model is defined by vectors associated with each state with a dimension of, say, 50, together with a global mapping from this vector space to the space of parameters of the GMM. This model appears to give better results than a conventional model, and the extra structure offers many new opportunities for modeling innovations while maintaining compatibility with most standard techniques.

show abstract

Section: Introductionmentioning

confidence: 99%

The subspace Gaussian mixture model—A structured model for speech recognition

Povey

Burget

Agarwal

et al. 2011

Computer Speech & Language

257

195

View full text Add to dashboard Cite

show abstract

“…There are 163 speakers (of both (sr) can be computed from Eqn. (9). As a result, the Q(w) function is quadratic and its derivative is linear, and the optimal weights can be found by solving a system of linear equation as expected.…”

Section: A1 Tidigits Corpusmentioning

confidence: 99%

“…Since the number of estimation parameters is greatly reduced, fast speaker adaptation using EV adaptation is possible with a few seconds of speech. The simple algorithm was later extended to work for large-vocabulary continuous speech recognition [9], [10], eigenspace-based MLLR [11], [12], and to approximate the model prior in MAP adaptation [13], [14], [15]. In addition, the eigenspace may be learned automatically by MLES [16], or during model training as in CAT [17].…”

Section: Introductionmentioning

confidence: 99%

Embedded kernel eigenvoice speaker adaptation and its implication to reference speaker weighting

Mak

Hsiao

et al. 2006

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-Recently, we proposed an improvement to the conventional eigenvoice (EV) speaker adaptation using kernel methods. In our novel kernel eigenvoice (KEV) speaker adaptation [1], speaker supervectors are mapped to a kernelinduced high dimensional feature space, where eigenvoices are computed using kernel principal component analysis. A new speaker model is then constructed as a linear combination of the leading eigenvoices in the kernel-induced feature space. KEV adaptation was shown to outperform EV, MAP, and MLLR adaptation in a TIDIGITS task with less than 10s of adaptation speech [2]. Nonetheless, due to many kernel evaluations, both adaptation and subsequent recognition in KEV adaptation are considerably slower than conventional EV adaptation. In this paper, we solve the efficiency problem and eliminate all kernel evaluations involving adaptation or testing observations by finding an approximate preimage of the implicit adapted model found by KEV adaptation in the feature space; we call our new method embedded kernel eigenvoice (eKEV) adaptation. eKEV adaptation is faster than KEV adaptation, and subsequent recognition runs as fast as normal HMM decoding.eKEV adaptation makes use of multi-dimensional scaling technique so that the resulting adapted model lies in the span of a subset of carefully chosen training speakers. It is related to the reference speaker weighting (RSW) adaptation method that is based on speaker clustering. Our experimental results on Wall Street Journal show that eKEV adaptation continues to outperform EV, MAP, MLLR, and the original RSW method. However, by adopting the way we choose the subset of reference speakers for eKEV adaptation, we may also improve RSW adaptation so that it performs as well as our eKEV adaptation.

show abstract

“…), and each individual speaker is then a point in the eigenspace. The simple algorithm was later extended to work for large-vocabulary continuous speech recognition [5,6,7]. In practice, a few to a few tens of eigenvoices are found adequate for fast speaker adaptation.…”

Section: Introductionmentioning

confidence: 99%

A study of various composite kernels for kernel eigenvoice speaker adaptation

Mak

Kwok

2004 IEEE International Conference on Acoustics, Speech, and Signal Processing

View full text Add to dashboard Cite

Eigenvoice-based methods have been shown to be effective for fast speaker adaptation when the amount of adaptation data is small, say, less than 10 seconds. In traditional eigenvoice (EV) speaker adaptation, linear principal component analysis (PCA) is used to derive the eigenvoices. Recently, we proposed that eigenvoices found by nonlinear kernel PCA could be more effective, and the eigenvoices thus derived were called kernel eigenvoices (KEV). One of our novelties is the use of composite kernel that makes it possible to compute state observation likelihoods via kernel functions. In this paper, we investigate two different composite kernels: direct sum kernel and tensor product kernel for KEV adaptation. In an evaluation on the TIDIGITS task, it is found that KEV speaker adaptation using both forms of composite kernel are equally effective, and they outperform a speaker-independent model and the adapted models from EV, MAP, or MLLR adaptation using 2.1s and 4.1s of speech. For example, with 2.1s of adaptation data, KEV adaptation outperforms the speaker-independent model by 27.5%, whereas EV, MAP, or MLLR adaptation are not effective at all.

show abstract

Very fast adaptation with a compact context-dependent eigenvoice model

Cited by 15 publications

References 8 publications

The subspace Gaussian mixture model—A structured model for speech recognition

The subspace Gaussian mixture model—A structured model for speech recognition

Embedded kernel eigenvoice speaker adaptation and its implication to reference speaker weighting

A study of various composite kernels for kernel eigenvoice speaker adaptation

Contact Info

Product

Resources

About