A Comparative Study of Feature and Score Normalization for Speaker Verification

Zheng, Rong; Zhang, Shuwu; Xu, Bo

doi:10.1007/11608288_71

Cited by 14 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CMVN reduces the effect of additive environmental noises from the feature vectors and it was previously shown that it improves the recognition rate in speaker recognition [33].…”

Section: Feature Normalizationmentioning

confidence: 98%

“…CVN has more impact on reducing the effect of additive environmental noise from feature vectors than CMN operation. However, in many speech and speaker recognition studies, CMN and CVN are combined which yields to have feature vectors with zero mean and unit variance which is known as cepstral mean and variance normalization (CMVN) [4,33]:…”

Section: Feature Normalizationmentioning

confidence: 99%

“…Another frequently used method to reduce the unwanted additive noise effect from feature vectors is the cepstral variance normalization (CVN) [33]. CVN normalizes the variance of each feature vector extracted from each speech frame to one, namely…”

Section: Feature Normalizationmentioning

confidence: 99%

See 2 more Smart Citations

Optimizing acoustic features for source cell-phone recognition using speech signals

Hanilçi

Ertaş

2013

Proceedings of the First ACM Workshop on Information Hiding and Multimedia Security

View full text Add to dashboard Cite

This paper presents comparison and optimization of acoustic features for source cell-phone recognition using recorded speech signals. Different acoustic feature extraction methods such as Mel-frequency, linear frequency and Bark frequency cepstral coefficients (MFCC, LFCC and BFCC) and linear prediction cepstral coefficients (LPCC) are considered. In addition to different feature sets, the effect of dynamic features, delta and double-delta coefficients (∆ and ∆ 2 ), and feature normalizations, cepstral mean normalization (CMN), cepstral variance normalization (CVN) and cepstral mean and variance normalization (CMVN) are also examined on the performance of source cell-phone recognition. The same support vector machine (SVM) classifier with fixed parameters and the same cell-phone dataset are used in the experiments in order to make a fair comparison of different features and feature normalization techniques.

show abstract

“…CMVN reduces the effect of additive environmental noises from the feature vectors and it was previously shown that it improves the recognition rate in speaker recognition [33].…”

Section: Feature Normalizationmentioning

confidence: 98%

Section: Feature Normalizationmentioning

confidence: 99%

See 1 more Smart Citation

Optimizing acoustic features for source cell-phone recognition using speech signals

Hanilçi

Ertaş

2013

Proceedings of the First ACM Workshop on Information Hiding and Multimedia Security

View full text Add to dashboard Cite

show abstract

“…2) Minimum Value of Detection Cost Function (minDCF) [14] Application scenarios have different requirements for false-alarm and miss rates. Thus, the setting of the threshold needs to be adjusted accordingly.…”

Section: Evaluation Measuresmentioning

confidence: 99%

Speaker Verification Based on Log-Likelihood Score Normalization

Cao¹,

Chen²,

Cao³

2020

JCC

View full text Add to dashboard Cite

Due to differences in the distribution of scores for different trials, the performance of a speaker verification system will be seriously diminished if raw scores are directly used for detection with a unified threshold value. As such, the scores must be normalized. To tackle the shortcomings of score normalization methods, we propose a speaker verification system based on log-likelihood normalization (LLN). Without a priori knowledge, LLN increases the separation between scores of target and non-target speaker models, so as to improve score aliasing of "same-speaker" and "different-speaker" trials corresponding to the same test speech, enabling better discrimination and decision capability. The experiment shows that LLN is an effective method of scoring normalization.

show abstract

“…The effect of the model normalisation can be observed in Figure 3 (f). This normalisation is called ZT-norm and in [11] a reduction of 20% EER is reported when compared with standard Z-Norm.…”

Section: Model Normalisationmentioning

confidence: 99%

Automatic Cross-Biometric Footstep Database Labelling Using Speaker Recognition

Vera-Rodríguez

Mason

Evans

2009

Advances in Biometrics

View full text Add to dashboard Cite

The often daunting task of collecting and manually labelling biometric databases can be a barrier to research. This is especially true for a new or non-established biometric such as footsteps. The availability of very large data sets often plays a role in the research of complex modelling and normalisation algorithms and so an automatic, semi-unsupervised approach to reduce the cost of manual labelling is potentially of immense value.This paper proposes a novel, iterative and adaptive approach to the automatic labelling of what is thought to be the first large scale footstep database (more than 10,000 examples across 127 persons). The procedure involves the simultaneous collection of a spoken, speaker-dependent password which is used to label the footstep data automatically via a pre-trained speaker recognition system. Subsets of labels are manually checked by listening to the particular password utterance, or viewing the associated talking face; both are recorded with the same time stamp as the footstep sequence.Experiments to assess the resulting label accuracy, based on manually labelled subsets, suggest that the accuracy of the automatic labelling is better than 0.1%, and thus sufficient to assess a biometric such as footsteps, which is anticipated to have a much higher error rate.

show abstract

A Comparative Study of Feature and Score Normalization for Speaker Verification

Cited by 14 publications

References 10 publications

Optimizing acoustic features for source cell-phone recognition using speech signals

Optimizing acoustic features for source cell-phone recognition using speech signals

Speaker Verification Based on Log-Likelihood Score Normalization

Automatic Cross-Biometric Footstep Database Labelling Using Speaker Recognition

Contact Info

Product

Resources

About