A noise robust i-vector extractor using vector taylor series for speaker recognition

Lei, Yun; Burget, Lukáš; Scheffer, Nicolas

doi:10.1109/icassp.2013.6638976

Cited by 59 publications

(36 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Various other methods aimed at solving this problem include using vector Taylor series (VTS) approximation (Lei et al 2013), multiple support vector machines (SVM) which are trained using adaptive boosting (Sarkar and Rao 2014), simplified VTS (sVTS) which reduces computational complexity of VTS (Lei et al 2014a), acoustic feature uncertainty propagation (Yu et al 2014), unscented transform which is used instead of VTS (Martinez et al 2014), etc.…”

Section: Noisy Datamentioning

confidence: 99%

i-Vectors in speech processing applications: a survey

Verma

Das

2015

Int J Speech Technol

View full text Add to dashboard Cite

In the domain of speech recognition many methods have been proposed over time like Gaussian mixture models (GMM), GMM with universal background model (GMM-UBM framework), joint factor analysis, etc. i-Vector subspace modeling is one of the recent methods that has become the state of the art technique in this domain. This method largely provides the benefit of modeling both the intra-domain and inter-domain variabilities into the same low dimensional space. In this survey, we present a comprehensive collection of research work related to i-vectors since its inception. Some recent trends of using i-vectors in combination with other approaches are also discussed. The application of i-vectors in various fields of speech recognition, viz speaker, language, accent recognition, etc. is also presented. This paper should serve as a good starting point for anyone interested in working with i-vectors for speech processing in general. We then conclude the paper with a brief discussion on the future of i-vectors.

show abstract

Section: Noisy Datamentioning

confidence: 99%

i-Vectors in speech processing applications: a survey

Verma

Das

2015

Int J Speech Technol

View full text Add to dashboard Cite

show abstract

“…This section describes the original idea of applying the VTS approximation to the model for noise robust i-vector extraction as introduced in [6]. The VTS-based i-vector extraction is a two-step process: 1) the UBM is first adapted to the additive and convolutive noise of a speech segment, and 2) the noise-compensated i-vector is then extracted based on the sufficient statistics collected from the adapted UBM.…”

Section: Vts-based I-vector Extractionmentioning

confidence: 99%

“…In our previous work [6], we proposed to tackle the problem at an earlier stage, where the i-vector extractor explicitly takes into account the modeling of noise in the speech data using the VTS approximation. VTS is used to model nonlinear distortions in the melcepstral domain caused by both additive and convolutive noise.…”

Section: Introductionmentioning

confidence: 99%

Simplified VTS-based I-vector extraction in noise-robust speaker recognition

Lei

McLaren

Ferrer

et al. 2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

A vector taylor series (VTS) based i-vector extractor was recently proposed for noise-robust speaker recognition by extracting synthesized clean i-vectors to be used in the standard system back-end. This approach brings significant improvements in accuracy for noisy speech conditions. However, this approach incurred such a large computational expense that using the state-of-the-art model size or evaluating large scale evaluations was impractical. In this work, we propose an efficient simplification scheme, named sVTS, in order to show that the VTS approach gives improvements in large scale applications compared to state-of-the-art systems. In contrast to VTS, sVTS generates normalized Baum-Welch statistics and uses the standard i-vector model, making it straightforward to employ on the state-of-the-art i-vector speaker recognition system. Results presented on both the PRISM and the large NIST SRE'12 corpora show that using sVTS i-vectors provides significant improvements in the noisy conditions, and that our proposed simplification result in only a slight degradation with respect to the original VTS approach.

show abstract

“…Further, the general approach of multi-condition training allows the model to capture observed noisy conditions (Garcia-Romero et al, 2012). Recently, well-known techniques widely used in robust automatic speech recognition, e.g., vector Taylor series (VTS), have been adopted for noise robust speaker recognition (Lei et al, 2013).…”

Section: Introductionmentioning

confidence: 99%