-Gaussian mixture modelling for speaker recognition

Wu, Dalei; Li, Ji; Wu, Haiqing

doi:10.1016/j.patrec.2008.12.013

Cited by 18 publications

(22 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For a useful review, refer to [78]. Another recent extension of GMM is based on nonlinear warping of the GMM density function [82]. These methods, however, lack formulation for the background model adaptation [68], which is an essential part of modern speaker verification relying on MAP training (Fig.…”

Section: Review Of Clustering Methods In Speaker Recognitionmentioning

confidence: 99%

Comparison of clustering methods: A case study of text-independent speaker modeling

Kinnunen¹,

Sidoroff²,

Tuononen³

et al. 2011

Pattern Recognition Letters

View full text Add to dashboard Cite

Clustering is needed in various applications such as biometric person authentication, speech coding and recognition, image compression and information retrieval. Hundreds of clustering methods have been proposed for the task in various fields but, surprisingly, there are few extensive studies actually comparing them. An important question is how much the choice of a clustering method matters for the final pattern recognition application. Our goal is to provide a thorough experimental comparison of clustering methods for text-independent speaker verification. We consider parametric Gaussian mixture model (GMM) and non-parametric vector quantization (VQ) model using the best known clustering algorithms including iterative (K-means, random swap, expectation-maximization), hierarchical (pairwise nearest neighbor, split, split-and-merge), evolutionary (genetic algorithm), neural (self-organizing map) and fuzzy (fuzzy C-means) approaches. We study recognition accuracy, processing time, clustering validity, and correlation of clustering quality and recognition accuracy. Experiments from these complementary observations indicate clustering is not a critical task in speaker recognition and the choice of the algorithm should be based on computational complexity and simplicity of the implementation. This is mainly because of three reasons: the data is not clustered, large models are used and only the best algorithms are considered. For low-order models, choice of the algorithm, however, can have a significant effect. Index Terms

show abstract

Section: Review Of Clustering Methods In Speaker Recognitionmentioning

confidence: 99%

Comparison of clustering methods: A case study of text-independent speaker modeling

Kinnunen¹,

Sidoroff²,

Tuononen³

et al. 2011

Pattern Recognition Letters

View full text Add to dashboard Cite

show abstract

“…In [19], the authors proposed the -integration of Gaussian densities as an extension of the classical GMM for speakers modeling in a speaker identification task. The -integration generalizes the linear combination adopted in the conventional GMM.…”

Section: B -Gmmmentioning

confidence: 99%

“…The essential idea was to improve the identification performance by emulating an integration process similar to what occurs inside a human brain. In [19], it was shown that the -GMM outperforms the conventional GMM in speaker identification task with speech transmitted through a fixed phone channel. In [18], the -GMM was applied to a speaker verification task in noisy conditions, and it also achieved better results than the GMM.…”

Section: B -Gmmmentioning

confidence: 99%

“…In [18], the -GMM was applied to a speaker verification task in noisy conditions, and it also achieved better results than the GMM. Given a speaker model , composed of Gaussian densities , , the -integration of the densities is defined as [19] (5) where are non-negative mixture weights constrained to , is a normalization constant, and is given by (6) From (6), the inverse of can be calculated by (7) The -GMM of speaker can be rewritten as (8) Note that the -integration of the Gaussian densities in (8) turns into a linear combination for , which corresponds to the conventional GMM. By choosing values of smaller than -1, the -GMM classifier emphasizes the larger probability values, and de-emphasizes the smaller ones.…”

Section: B -Gmmmentioning

confidence: 99%

“…In this paper, the MT based on colored (Colored-MT) [17], white (White-MT) and narrow-band (Narrow-MT) [8] noises are applied with the -GMM [19] classifier to improve the noise robustness of speaker verification. The conventional GMM is considered as a particular case of the -GMM ( ).…”

mentioning

confidence: 99%

See 2 more Smart Citations

On speech features fusion, α-integration Gaussian modeling and multi-style training for noise robust speaker classification

Venturini

Zão

Coelho

2014

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

This paper investigates the fusion of Mel-frequency cepstral coefficients (MFCC) and statistical pH features to improve the performance of speaker verification (SV) in non-stationary noise conditions. The -integrated Gaussian Mixture Model ( -GMM) classifier is adopted for speaker modeling. Two different approaches are applied to reduce the effects of noise corruption in the SV task: speech enhancement and multi-style training (MT). The spectral subtraction with minimum statistics (MS/SS) and the optimally-modified log-spectral amplitude with improved minima controlled recursive averaging (IMCRA/OMLSA) are examined for the speech enhancement procedure. The MT techniques are based on colored (Colored-MT), white (White-MT) and narrow-band (Narrow-MT) noises. Six real non-stationary noises, collected from different acoustic sources, are used to corrupt the TIMIT speech database in four different signal-to-noise ratios (SNR). The index of non-stationarity (INS) is chosen for the stationarity tests of the acoustic noises. Complementary SV experiments are conducted in realistic noisy conditions using the MIT database. The results show that the best SV accuracy was obtained with the MFCC + pH features fusion, the MS/SS and the Colored-MT.Index Terms-Features fusion, Hurst exponent, multi-style training, non-stationary acoustic noise, speaker verification, speech enhancement, -GMM.

show abstract

Bionic optimization of MFCC features based on speaker fast recognition

Lin

Chen

2021

Applied Acoustics

View full text Add to dashboard Cite

-Gaussian mixture modelling for speaker recognition

Cited by 18 publications

References 9 publications

Comparison of clustering methods: A case study of text-independent speaker modeling

Comparison of clustering methods: A case study of text-independent speaker modeling

On speech features fusion, α-integration Gaussian modeling and multi-style training for noise robust speaker classification

Bionic optimization of MFCC features based on speaker fast recognition

Contact Info

Product

Resources

About