Within-class covariance normalization for SVM-based speaker recognition

Hatch, Andrew; Kajarekar, Sachin; Stolcke, Andreas

doi:10.21437/interspeech.2006-183

Cited by 249 publications

(18 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For phrase-dependent TD-SV, it is preferable to train the embedding extractor on the matched phrases to make the speaker embeddings reflect the phonetic variability in the pre-defined phrases. Also, the parameters in the conventional channel compensation methods-such as within-class covariance normalization (WCCN) [31], LDA, and PLDA-should be trained on the data with the predefined phrases. This strategy helps reject impostors speaking the wrong phrases [32].…”

Section: B Text-dependent Speaker Verificationmentioning

confidence: 99%

A Survey on Text-Dependent and Text-Independent Speaker Verification

Lin

Mak

2022

IEEE Access

View full text Add to dashboard Cite

Speaker verification (SV) aims to detect an individual's identity from his/her voice. SV has been successfully applied in various areas such as access control, remote service customization, financial transactions, etc. Depending on whether the text content is pre-defined or not, SV can be text-dependent or text-independent. This paper reviews recent research on text-dependent SV (TD-SV) and text-independent SV (TI-SV). Because most modern SV systems apply deep learning methods to boost performance, we focus on the studies that use deep speaker embedding, a technique representing a person's identity via a fixed-dimensional vector encoded from a variable-length utterance. Rather than detailing every existing SV system, we make an overview of the representative SV systems that have attracted wide attention. Furthermore, an increasing number of SV systems have been devoted to addressing real-world challenges such as reverberation and noise, and this has driven a large number of studies on practical SV. Therefore, the survey compares the existing SV systems in the Far-Field Speaker Verification Challenge 2020 (FFSVC 2020) to illustrate the most effective techniques for both TD-SV and TI-SV. INDEX TERMSText-dependent speaker verification, text-independent speaker verification, deep speaker embedding, far-field speaker verification

show abstract

Section: B Text-dependent Speaker Verificationmentioning

confidence: 99%

A Survey on Text-Dependent and Text-Independent Speaker Verification

Lin

Mak

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…We followed the standard procedure to pre-process the ivectors for Gaussian PLDA modeling. Specifically, the 500dimensional senone i-vectors were whitened by within-class covariance normalization (WCCN) [40] and length normalization [29], followed by linear discriminant analysis to reduce the dimension to 200 and variance normalization by WCCN [41]. These 200-dimensional i-vectors were input to the PLDA model and the DNNs.…”

Section: Denoised Senone I-vectorsmentioning

confidence: 99%

DNN-Based Score Calibration With Multitask Learning for Noise Robust Speaker Verification

Tan

Mak

2018

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

This paper proposes and investigates several deep neural network (DNN)-based score compensation, transformation and calibration algorithms for enhancing the noise robustness of i-vector speaker verification systems. Unlike conventional calibration methods where the required score shift is a linear function of SNR or log-duration, the DNN approach learns the complex relationship between the score shifts and the combination of i-vector pairs and uncalibrated scores. Furthermore, with the flexibility of DNNs, it is possible to explicitly train a DNN to recover the clean scores without having to estimate the score shifts. To alleviate the overfitting problem, multi-task learning is applied to incorporate auxiliary information such as SNRs and speaker ID of training utterances into the DNN. Experiments on NIST 2012 SRE show that score calibration derived from multi-task DNNs can improve the performance of the conventional score-shift approch significantly, especially under noisy conditions.

show abstract

“…where ω is the global mean of i-vectors, W is a transformation matrix obtained from the Cholesky decomposition of the within-class covariance matrix of ivectors [18] and ω wht is the whitened i-vector. The second step is to apply a simple length-normalization to the whitened i-vectors:…”

Section: Pre-processing For Gaussian Pldamentioning

confidence: 99%

“…It is customary to include linear discriminant analysis (LDA) and within-class covariance normalization (WCCN) [18] in the pre-processing steps. The whole pre-processing can be written in a more succinct fashion:…”

Section: Pre-processing For Gaussian Pldamentioning

confidence: 99%