Background model design for flexible and portable speaker verification systems

Siohan, Olivier; Lee, Chin‐Hui; Surendran, Arun C.; Li, Qi

doi:10.1109/icassp.1999.759798

Cited by 15 publications

(9 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For comparison purposes, the proposed PLR test is compared with the other three algorithms, which are HMM-based and utilize only meaningful patterns. These are the LRT algorithm based on cohort (LRT-cohort) [19], the LRT algorithm based on perturbed background models (LRT-PBM) [20] and THMM algorithm [11].…”

Section: Resultsmentioning

confidence: 99%

“…Reynolds et al [15] used patterns from a set of meaningful classes to train a unique background model based on the Gaussian mixture model (GMM) which they called universal background model (UBM). Siohan et al [20] proposed a perturbed training method to train a specific background model for each meaningful class using their own patterns. The background models are also HMMs that are trained with the same training samples used in activity HMMs.…”

Section: Review Of Related Workmentioning

confidence: 99%

“…It is difficult to establish these alternative models in activity recognition since non-meaningful activities may be similar to meaningful activities and their possibilities are infinite. A few approaches have also been proposed to build alternative models using only meaningful samples [7,20]. However, their approximated alternative models are not guaranteed to be proper.…”

Section: Introductionmentioning

confidence: 98%

See 2 more Smart Citations

Rejection of non-meaningful activities for HMM-based activity recognition system

Liu

Chua

2010

Image and Vision Computing

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

Section: Review Of Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Rejection of non-meaningful activities for HMM-based activity recognition system

Liu

Chua

2010

Image and Vision Computing

View full text Add to dashboard Cite

“…In such an SV system, it is desirable to perform SV using only utterances of enrolled speakers. This issue was already addressed in text-dependent SV using hidden Markov model [4] and text-independent SV using GMM [5], [6].…”

Section: Introductionmentioning

confidence: 99%

Artificial Cohort Generation Based on Statistics of Real Cohorts for GMM-Based Speaker Verification

Mukai

Noda

Osanai

2011

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThis paper discusses speaker verification (SV) using Gaussian mixture models (GMMs), where only utterances of enrolled speakers are required. Such an SV system can be realized using artificially generated cohorts instead of real cohorts from speaker databases. This paper presents a rational approach to set GMM parameters for artificial cohorts based on statistics of GMM parameters for real cohorts. Equal error rates for the proposed method are about 10% less than those for the previous method, where GMM parameters for artificial cohorts were set in an ad hoc manner.

show abstract

“…To improve the competitiveness of this model, a previous study [11] proposed the use of a modified normalizing model (MNM) determined by perturbing the inferred background model using the enrollment data to reflect the lexical content of the speaker's password. In another study [12], the authors proposed the use of the speaker enrollment data to (1) train a background model with fewer number of parameters compared to the speaker model or (2) perturbing the temporal information by reversing the state order of the previously trained background model. In the work reported here, we will demonstrate that the use of multiple background models, corresponding to the inferred speakerindependent password HMM models can improve the performance of the UCP-SV system.…”

Section: Speaker Verification 71 Score Normalizationmentioning

confidence: 99%

User-customized password speaker verification using multiple reference and background models

BenZeghiba

Bourlard

2006

Speech Communication

View full text Add to dashboard Cite

Abstract. This paper discusses and optimizes an HMM/GMM based User-Customized Password Speaker Verification (UCP-SV) system. Unlike text-dependent speaker verification, in UCP-SV systems, customers can choose their own passwords with no lexical constraints. The password has to be pronounced a few times during the enrollment step to create a customer dependent model. Although potentially more "user-friendly", such systems are less understood and actually exhibit several practical issues, including automatic HMM inference, speaker adaptation, and efficient likelihood normalization. In our case, HMM inference (HMM topology) is performed using hybrid HMM/MLP systems, while the parameters of the inferred model, as well as their adaptation, will use GMMs. However, the evaluation of a UCP-SV baseline system shows that the background model used for likelihood normalization is the main difficulty. Therefore, to circumvent this problem, the main contribution of the paper is to investigate the use of multiple reference models for customer acoustic modeling and multiple background models for likelihood normalization. In this framework, several scoring techniques are investigated, such as Dynamic Model Selection (DMS) and fusion techniques. Results on two different experimental protocols show that an appropriate selection criteria for customer and background models can improve significantly the UCP-SV performance, making the UCP-SV system quite competitive with a text-dependent SV system. Finally, as customers' passwords are short, a comparative experiment using the conventional GMM-UBM text-independent approach is also conducted. 2 IDIAP-RR 04-41

show abstract

Background model design for flexible and portable speaker verification systems

Cited by 15 publications

References 4 publications

Rejection of non-meaningful activities for HMM-based activity recognition system

Rejection of non-meaningful activities for HMM-based activity recognition system

Artificial Cohort Generation Based on Statistics of Real Cohorts for GMM-Based Speaker Verification

User-customized password speaker verification using multiple reference and background models

Contact Info

Product

Resources

About