Comparing spectrum estimators in speaker verification under additive noise degradation

Hanilçi, Cemal; Kinnunen, Tomi; Saeidi, Rahim; Pohjalainen, Jouni; Alku, Paavo; Ertaş, Figen; Sandberg, Johan; Hansson-Sandsten, Maria

doi:10.1109/icassp.2012.6288985

Cited by 15 publications

(12 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At feature level, [3] carried out an extensive comparison of several spectrum estimation methods under additive noise contamination and found that the best spectrum estimator was related to the noise type and level. Recent work [4,5], based on vector Taylor series (VTS) then developed using "unscented transforms" [6] tried to model non-linear distortions in the cepstral domain based on a non-linear noise model in order to relate clean and noisy cepstral coefficients and help estimate a "cleaned-up" version of i-vectors.…”

Section: Introductionmentioning

confidence: 99%

Fast i-vector denoising using MAP estimation and a noise distributions database for robust speaker recognition

Kheder

Matrouf

Bousquet

et al. 2017

Computer Speech & Language

View full text Add to dashboard Cite

Once the i-vector paradigm has been introduced in the field of speaker recognition, many techniques have been proposed to deal with additive noise within this framework. Due to the complexity of its effect in the i-vector space, a lot of effort has been put into dealing with noise in other domains (speech enhancement, feature compensation, robust i-vector extraction and robust scoring). As far as we know, there was no serious attempt to handle the noise problem directly in the i-vector space without relying on data distributions computed on a prior domain. The aim of this paper is twofold. First, it proposes a fullcovariance Gaussian modeling of the clean i-vectors and noise distribution in the i-vector space and introduces a technique to estimate a clean i-vector given the noisy version and the noise density function using the MAP approach. Based on NIST data, we show that it is possible to improve by up to 60% the baseline system performance. Second, in order to make this algorithm usable in a real application and reduce the computational time needed by i-MAP, we propose an extension that requires building a noise distribution database in the i-vector space in an off-line step and using it later in the test phase. We show that it is possible to achieve comparable results using this approach (up to 57% of relative EER improvement) with a sufficiently large noise distribution database.

show abstract

Section: Introductionmentioning

confidence: 99%

Fast i-vector denoising using MAP estimation and a noise distributions database for robust speaker recognition

Kheder

Matrouf

Bousquet

et al. 2017

Computer Speech & Language

View full text Add to dashboard Cite

show abstract

“…At a feature level, [3] carried out an extensive comparison of several spectrum estimation methods and found that the best estimator was related to the noise type and SNR level. Recent work [4,5], based on vector Taylor series (VTS) then developed using "unscented transforms" [6] tried to model non-linear distortions in the cepstral domain based on a nonlinear noise model in order to relate clean and noisy cepstral coefficients and help estimate a "cleaned-up" version of ivectors.…”

Section: Introductionmentioning

confidence: 99%

Additive noise compensation in the i-vector space for speaker recognition

Kheder

Matrouf

Bonastre

et al. 2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

State-of-the-art speaker recognition systems performance degrades considerably in noisy environments even though they achieve very good results in clean conditions. In order to deal with this strong limitation, we aim in this work to remove the noisy part of an i-vector directly in the i-vector space. Our approach offers the advantage to operate only at the i-vector extraction level, letting the other steps of the system unchanged. A maximum a posteriori (MAP) procedure is applied in order to obtain clean version of the noisy i-vectors taking advantage of prior knowledge about clean i-vectors distribution. To perform this MAP estimation, Gaussian assumptions over clean and noise i-vectors distributions are made. Operating on NIST 2008 data, we show a relative improvement up to 60% compared with baseline system. Our approach also outperforms the "multi-style" backend training technique. The efficiency of the proposed method is obtained at the price of relative high computational cost. We present at the end some ideas to improve this aspect.

show abstract

“…The resulting optimal weights for two different cases of ρ are presented where the corresponding covariance matrix R x is used in Equation 20. The AR(1) spectrum is a simple model but reasonable for speech data as speech data often are estimated as AR models (order [10][11][12][13][14][15][16][17][18][19][20]. The average damping of the different poles (ρ) of such an estimated AR spectrum from real data will give an idea of what damping factor should be chosen for the AR(1) model for the optimization of the weights.…”

Section: Variance Of the Log-spectrummentioning

confidence: 99%

“…This estimator has been evaluated and compared with the Thomson multitapers, the sinusoidal multitapers, the Welch method, and usual windowed periodogram-based cepstrum analysis methods for speaker recognition. The results of these studies show that a multitaper estimator optimal for a speech-like spectrum model has advantages compared to traditional techniques [14][15][16].…”

Section: Introductionmentioning

confidence: 99%

Mean square error optimal weighting for multitaper cepstrum estimation

Hansson-Sandsten

2013

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

The aim of this paper is to find a multitaper-based spectrum estimator that is mean square error optimal for cepstrum coefficient estimation. The multitaper spectrum estimator consists of windowed periodograms which are weighted together, where the weights are optimized using the Taylor expansion of the log-spectrum variance and a novel approximation for the log-spectrum bias. A thorough discussion and evaluation are also made for different bias approximations for the log-spectrum of multitaper estimators. The optimized weights are applied together with the sinusoidal tapers as the multitaper estimator. Comparisons of the cepstrum mean square error are made of some known multitaper methods as well as with the parametric autoregressive estimator for simulated speech signals.

show abstract

Comparing spectrum estimators in speaker verification under additive noise degradation

Cited by 15 publications

References 16 publications

Fast i-vector denoising using MAP estimation and a noise distributions database for robust speaker recognition

Fast i-vector denoising using MAP estimation and a noise distributions database for robust speaker recognition

Additive noise compensation in the i-vector space for speaker recognition

Mean square error optimal weighting for multitaper cepstrum estimation

Contact Info

Product

Resources

About