Nonnegative matrix factorization based noise robust speaker verification

Liu, S. H.; Zou, Yuexian; Ning, Hongke

doi:10.1109/chinasip.2015.7230357

Cited by 2 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the first class of algorithms, spectral and wavelet-based speech enhancement techniques were studied in [1,2] and were proven to be noise and SNR level-dependent (performance can improve or degrade depending on the noise / SNR). NFM-based speech enhancement algorithms were also presented in [3][4][5] and showed relatively low improvement compared to other techniques (10% of relative improvement in EER. In the cepstral domain, several stochastic compensation techniques were applied in [6] (such as RATZ [7], SPLICE [8], TRAJMAP [9] and SSM [10]) and were proven to be highly efficient in noisy environments but such algorithms assume prior knowledge about the test noise.…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Approach Using Joint Clean and Noisy i-Vectors Modeling for Speaker Recognition

et al. 2016

View full text Add to dashboard Cite

Additive noise is one of the main challenges for automatic speaker recognition and several compensation techniques have been proposed to deal with this problem. In this paper, we present a new "data-driven" denoising technique operating in the i-vector space based on a joint modeling of clean and noisy i-vectors. The joint distribution is estimated using a large set of i-vectors pairs (clean i-vectors and their noisy versions generated artificially) then integrated in an MMSE estimator in the test phase to compute a "cleaned-up" version of noisy test ivectors. We show that this algorithm achieves up to 80% of relative improvement in EER. We also present a version of the proposed algorithm that can be used to compensate multiple "unseen" noises. We test this technique on the recently published SITW database and show a significant gain compared to the baseline system performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Probabilistic Approach Using Joint Clean and Noisy i-Vectors Modeling for Speaker Recognition

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Dealing with noise has also been one of the principal areas of interest and different techniques have been proposed to deal with it in different domains. In the temporal domain, speech enhancement techniques [11][12][13] have been proven to be noise and SNR-level dependent yielding low improvement rates (10% of relative EER improvement). In the cepstral domain, several stochastic compensation algorithms have been proposed in [14] and have been shown to be effective when prior knowledge about test noise is available.…”

Section: Introductionmentioning

confidence: 99%

LIA System for the SITW Speaker Recognition Challenge

et al. 2016

View full text Add to dashboard Cite

This paper presents the speaker verification systems developed in the LIA lab at the University of Avignon for the SITW (Speakers In The Wild) challenge. We present the algorithms used to deal with additive noise, short utterances and propose an improved scoring scheme using a discriminative classifier and integrating the homogeneity of the two compared recordings. Due to the heterogeneity of this database (presence of background noise, reverberation, Lombard effect, etc.), it is hard to analyze the contribution of individual techniques used to deal with each problem. For this reason, a subset of the trials will be studied for each algorithm in order to emphasize its contribution.

show abstract

Nonnegative matrix factorization based noise robust speaker verification

Cited by 2 publications

References 19 publications

Probabilistic Approach Using Joint Clean and Noisy i-Vectors Modeling for Speaker Recognition

Probabilistic Approach Using Joint Clean and Noisy i-Vectors Modeling for Speaker Recognition

LIA System for the SITW Speaker Recognition Challenge

Contact Info

Product

Resources

About