Maximum likelihood modeling with Gaussian distributions for classification

Gopinath, R.

doi:10.1109/icassp.1998.675351

Cited by 222 publications

(102 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The number of classes was 43, corresponding to the number of monophones. MLLT [15] was applied after LDA, HDA, ODA and HLDAC. For PLDA, we assumed that projected class covariance matrices in Eq.…”

Section: Feature Transformation Proceduresmentioning

confidence: 99%

Evaluation of Combinational Use of Discriminant Analysis-Based Acoustic Feature Transformation and Discriminative Training

Sakai

Kitaoka

Hattori

et al. 2010

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYTo improve speech recognition performance, acoustic feature transformation based on discriminant analysis has been widely used. For the same purpose, discriminative training of HMMs has also been used. In this letter we investigate the effectiveness of these two techniques and their combination. We also investigate the robustness of matched and mismatched noise conditions between training and evaluation environments.

show abstract

Section: Feature Transformation Proceduresmentioning

confidence: 99%

Evaluation of Combinational Use of Discriminant Analysis-Based Acoustic Feature Transformation and Discriminative Training

Sakai

Kitaoka

Hattori

et al. 2010

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…The two transformed feature vectors, u t and v t , are concatenated for feature-level association. Since continuous HMMs having Gaussian mixture models with diagonal covariance matrices are used for recognition, the concatenated features are transformed further by using the maximum likelihood linear transform (MLLT) [18] so that each component of the feature vector is uncorrelated with each other.…”

Section: Feature-level Associationmentioning

confidence: 99%

Two-Level Bimodal Association for Audio-Visual Speech Recognition

Lee

Ebrahimi

2009

Advanced Concepts for Intelligent Vision Systems

View full text Add to dashboard Cite

Abstract. This paper proposes a new method for bimodal information fusion in audio-visual speech recognition, where cross-modal association is considered in two levels. First, the acoustic and the visual data streams are combined at the feature level by using the canonical correlation analysis, which deals with the problems of audio-visual synchronization and utilizing the cross-modal correlation. Second, information streams are integrated at the decision level for adaptive fusion of the streams according to the noise condition of the given speech datum. Experimental results demonstrate that the proposed method is effective for producing noise-robust recognition performance without a priori knowledge about the noise conditions of the speech data.

show abstract

“…Linear discriminant analysis (LDA) [13] for instance is a standard technique to minimize intra-class discriminability, to maximize inter-class discriminability and to extract relevant informations from high-dimensional features spanning larger contexts. Maximum likelihood linear transforms (MLLT) [14], [15] is a common method to de-correlate feature components, and feature-space maximum likelihood linear regression (fM-LLR) [16], [17] is widely used for speaker adaptation. Class discriminating properties are critical for clustering methods like the one used in [12], and adaptive feature transformations can help reduce variability.…”

Section: Introductionmentioning

confidence: 99%

Learning Supervised Feature Transformations on Zero Resources for Improved Acoustic Unit Discovery

Heck

Sakti

Nakamura

2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this work we utilize feature transformations that are common in supervised learning without having prior supervision, with the goal to improve Dirichlet process Gaussian mixture model (DPGMM) based acoustic unit discovery. The motivation of using such transformations is to create feature vectors that are more suitable for clustering. The need of labels for these methods makes it difficult to use them in a zero resource setting. To overcome this issue we utilize a first iteration of DPGMM clustering to generate frame based class labels for the target data. The labels serve as basis for learning linear discriminant analysis (LDA), maximum likelihood linear transform (MLLT) and feature-space maximum likelihood linear regression (fMLLR) based feature transformations. The novelty of our approach is the way how we use a traditional acoustic model training pipeline for supervised learning to estimate feature transformations in a zero resource scenario. We show that the learned transformations greatly support the DPGMM sampler in finding better clusters, according to the performance of the DPGMM posteriorgrams on the ABX sound class discriminability task. We also introduce a method for combining posteriorgram outputs of multiple clusterings and demonstrate that such combinations can further improve sound class discriminability.

show abstract

Maximum likelihood modeling with Gaussian distributions for classification

Cited by 222 publications

References 4 publications

Evaluation of Combinational Use of Discriminant Analysis-Based Acoustic Feature Transformation and Discriminative Training

Evaluation of Combinational Use of Discriminant Analysis-Based Acoustic Feature Transformation and Discriminative Training

Two-Level Bimodal Association for Audio-Visual Speech Recognition

Learning Supervised Feature Transformations on Zero Resources for Improved Acoustic Unit Discovery

Contact Info

Product

Resources

About