The Contribution of Cepstral and Stylistic Features to SRI's 2005 NIST Speaker Recognition Evaluation System

Ferrer, Luciana; Shriberg, Elizabeth; Kajarekar, Sachin; Stolcke, Andreas; Sönmez, Kemal; Venkataraman, Anand; Bratt, Harry

doi:10.1109/icassp.2006.1659967

Cited by 15 publications

(12 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…log(1 + e y i w t x i ) (2) In this paper, we consider a modified version of this problem, in which we allow weights to be applied to each term in the sum. This approach is used to compensate for the priors observed in the training data if those priors are expected to be different from the priors that will be found in the test data.…”

Section: Standard Linear Logistic Regressionmentioning

confidence: 99%

“…For lack of space we do not list references for every system here. Please refer to [2] for such a list. Cepstral GMM system: (0.276/6.15) This is a conventional cepstral Gaussian mixture model system adapted from a universal background model, using a 2048-component GMM.…”

Section: Individual Systemsmentioning

confidence: 99%

“…Score-level combination produces a final score that is later thresholded to obtain a decision. Examples of systems for which a score-level combination has resulted in significant gains over the best individual system can be found in [1,2,3,4,5]. Common combination procedures include neural networks [1,2], support vector machines (SVM) [2], linear logistic regression [3], and weighted summation using empirically determined weights [5].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

System combination using auxiliary information for speaker verification

Ferrer

Graciarena

Zymnis

et al. 2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

Recent studies in speaker recognition have shown that scorelevel combination of subsystems can yield significant performance gains over individual subsystems. We explore the use of auxiliary information to aid the combination procedure. We propose a modified linear logistic regression procedure that conditions combination weights on the auxiliary information. A regularization procedure is used to control the complexity of the extended model. Several auxiliary features are explored. Results are presented for data from the 2006 NIST speaker recognition evaluation (SRE). When an estimated degree of nonnativeness for the speaker is used as auxiliary information, the proposed combination results in a 15% relative reduction in equal error rate over methods based on standard linear logistic regression, support vector machines, and neural networks.

show abstract

Section: Standard Linear Logistic Regressionmentioning

confidence: 99%

Section: Individual Systemsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

System combination using auxiliary information for speaker verification

Ferrer

Graciarena

Zymnis

et al. 2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

show abstract

“…Their brief description can be found in [10] and they are listed in Table 1. Basically, the systems can be categorized either into acoustic (systems 1 to 3) or stylistic (systems 4 to 7) oriented.…”

Section: Dt DImentioning

confidence: 99%

Automatically Correcting Bias in Speaker Recognition Systems

Solewicz

Koppel

2006

2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing

View full text Add to dashboard Cite

In this paper we present a general machine learning framework for score bias reduction and analysis in speaker recognition systems. The general principle is to learn a meta-system using recognition systems' errors, given the training and testing conditions in which they occurred. In the context of speaker recognition, the proposed method is able to reduce the bias introduced in scores due to a variety of factors such as channel mismatch, additive noise, gender mismatch, different speaking styles, etc. Moreover, this framework enables a deep understanding of the origins of score bias in any system, which will support an optimized system redesign. Preliminary results obtained with several state-of-the-art systems showed considerable improvement in original performance, in addition to identifying sources of system bias.

show abstract

“…If the classifier outputs offer complementary information, classifier fusion can improve the performance. A number of fusion techniques with considerable improvements over a single system have been proposed in speaker and spoken language recognition based on linear score weighting [20], GMM [18], SVM [21], ANN [22], etc. in which the optimized weighting coefficients are applied to the scores produced by individual classifiers.…”

Section: Introductionmentioning

confidence: 99%

Error Corrective Fusion of Classifier Scores for Spoken Language Recognition

Dehzangi

Chng

et al. 2011

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThis paper investigates a new method for fusion of scores generated by multiple classification sub-systems that help to further reduce the classification error rate in Spoken Language Recognition (SLR). In recent studies, a variety of effective classification algorithms have been developed for SLR. Hence, it has been a common practice in the National Institute of Standards and Technology (NIST) Language Recognition Evaluations (LREs) to fuse the results from several classification sub-systems to boost the performance of the SLR systems. In this work, we introduce a discriminative performance measure to optimize the performance of the fusion of 7 language classifiers developed as IIR's submission to the 2009 NIST LRE. We present an Error Corrective Fusion (ECF) method in which we iteratively learn the fusion weights to minimize error rate of the fusion system. Experiments conducted on the 2009 NIST LRE corpus demonstrate a significant improvement compared to individual sub-systems. Comparison study is also conducted to show the effectiveness of the ECF method.

show abstract

The Contribution of Cepstral and Stylistic Features to SRI's 2005 NIST Speaker Recognition Evaluation System

Cited by 15 publications

References 10 publications

System combination using auxiliary information for speaker verification

System combination using auxiliary information for speaker verification

Automatically Correcting Bias in Speaker Recognition Systems

Error Corrective Fusion of Classifier Scores for Spoken Language Recognition

Contact Info

Product

Resources

About