Experiments in Session Variability Modelling for Speaker Verification

Vogt, Robbie; Sridharan, Sridha

doi:10.1109/icassp.2006.1660166

Cited by 29 publications

(37 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The factor analysis [6,7,8] has been found to be very effective in reducing the channel bias in the Gaussian mixture models and universal background models (GMM-UBM) system [6]. In this paper, we use factor analysis to reduce channel bias.…”

Section: Introductionmentioning

confidence: 99%

Double Gauss Based Unsupervised Score Normalization in Speaker Verification

Guo

Dai

Wang

2008

2008 6th International Symposium on Chinese Spoken Language Processing

View full text Add to dashboard Cite

In text-independent speaker verification, unsupervised mode can improve system performance. In traditional systems, the speaker model is updated when a test speech has a score higher than a particular threshold; we call this unsupervised model training. In this paper, an unsupervised score normalization is proposed. A target speaker score Gauss and an impostor score Gauss are set up as a prior; the parameters of the impostor score model are updated using the test score. Then the test score is normalized by the new impostor score model. When the unsupervised score normalization, unsupervised model training and factor analysis are adopted in the NIST 2006 SRE core test, the EER of the system is 4.29%.

show abstract

Section: Introductionmentioning

confidence: 99%

Double Gauss Based Unsupervised Score Normalization in Speaker Verification

Guo

Dai

Wang

2008

2008 6th International Symposium on Chinese Spoken Language Processing

View full text Add to dashboard Cite

show abstract

“…Direct modelling of session effects has led to a significant increase in robustness to channel and session variations in GMM-based speaker verification systems. Results show a reduction of 46% in EER and 42% in minimum detection cost over baseline GMM-UBM performance on the Mixer corpus of conversational telephony data [5].…”

Section: Introductionmentioning

confidence: 95%

“…Attempts to directly model session variability in GMM-UBM based speaker verification systems have provided significant performance improvements when using telephony speech [5]. The purpose of session variability modelling is to introduce a constrained offset of the speaker's mean vectors to represent the effects introduced by the session conditions.…”

Section: Session Variability Modellingmentioning

confidence: 99%

“…An efficient procedure for the optimisation of the model parameters is described in [5]. The session variability vectors are not actually retained to model the speaker but their estimation is necessary to accurately estimate the true speaker means.…”

Section: Session Variability Modellingmentioning

confidence: 99%

“…Recently, a method was proposed to directly model session variability in telephony speech during model training and testing [4,5]. The main assumption is that session effects can be represented as a set of offsets from the true speaker model means.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SVM Speaker Verification Using Session Variability Modelling and GMM Supervectors

McLaren

Vogt

Sridharan

Advances in Biometrics

Self Cite

View full text Add to dashboard Cite

Abstract. This paper demonstrates that modelling session variability during GMM training can improve the performance of a GMM supervector SVM speaker verification system. Recently, a method of modelling session variability in GMM-UBM systems has led to significant improvements when the training and testing conditions are subject to session effects. In this work, session variability modelling is applied during the extraction of GMM supervectors prior to SVM speaker model training and classification. Experiments performed on the NIST 2005 corpus show major improvements over the baseline GMM supervector SVM system.

show abstract