Regularized All-Pole Models for Speaker Verification Under Noisy Environments

Hanilçi, Cemal; Kinnunen, Tomi; Ertaş, Figen; Saeidi, Rahim; Pohjalainen, Jouni; Alku, Paavo

doi:10.1109/lsp.2012.2184284

Cited by 25 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several techniques are proposed to deal with this problem. In [17], regularized linear prediction (RLP) is applied to reduce the mismatch between training and test samples. RLP is a parametric spectral modelling that produces smooth spectra without changing the formant positions by penalizing rapid changes in all-pole spectral envelopes [84].…”

Section: Robust Features Against Additive Noisementioning

confidence: 99%

A Review on Feature Extraction for Speaker Recognition under Degraded Conditions

DİŞKEN

Tüfekçi

Sarıbulut

et al. 2016

IETE Technical Review

View full text Add to dashboard Cite

Speech is a signal that includes speaker's emotion, characteristic specification, phonemeinformation etc. Various methods have been proposed for speaker recognition by extracting specifications of a given utterance. Among them, short-term cepstral features are used excessively in speech, and speaker recognition areas because of their low complexity, and high performance in controlled environments. On the other hand, their performances decrease dramatically under degraded conditions such as channel mismatch, additive noise, emotional variability, etc. In this paper, a literature review on speaker-specific information extraction from speech is presented by considering the latest studies offering solutions to the aforementioned problem. The studies are categorized in three groups considering their robustness against channel mismatch, additive noise, and other degradations such as vocal effort, emotion mismatch, etc. For a more understandable representation, they are also classified into two tables by utilizing their classification methods, and used data-sets.

show abstract

Section: Robust Features Against Additive Noisementioning

confidence: 99%

A Review on Feature Extraction for Speaker Recognition under Degraded Conditions

DİŞKEN

Tüfekçi

Sarıbulut

et al. 2016

IETE Technical Review

View full text Add to dashboard Cite

show abstract

“…The phase normalization method is proposed in [9] to address the problem of change in phase according to the frame position in input speech. For speaker verification, regularized linear prediction based MFCC extraction is considered in [6]. The multitaper MFCC features for speaker verification [5] using ivectors are studied in [3].…”

Section: Related Workmentioning

confidence: 99%

“…MFCCs are obtained by taking Fourier transform of windowed speech frames [6]. In training phase, the features are computed by feature extraction scheme and then the speaker model is created from the features.…”

Section: Introductionmentioning

confidence: 99%

Combining dynamic features with MFCC for text-independent speaker identification

Chaudhari

Rahulkar

Dhonde

2015

2015 International Conference on Information Processing (ICIP)

View full text Add to dashboard Cite

This paper presents text-independent speaker identification scheme based on the combination of dynamic features with Mel Frequency Cepstral Coefficients (MFCC). First, MFCCs are used to extract the speaker specific features and then vector quantization is used to model the speaker. We compare the performance of speaker identification system using number of MFCC filters and centroids of vector quantization. In addition, we also compare the effect of possible combinations of MFCC with log energy, delta and double delta coefficients for number of filters and centroids of vector quantization on recognition rate. The experimental results have been evaluated on the developed database of 75 speakers.

show abstract

“…Also, since the MFCCs are widely adopted, many researchers have made effort to improve its robustness under noise by modifying, or changing, some processes in the conventional scheme [9][10][11][12][13]. Interested readers may refer to [14] for the recent progress in the feature extraction techniques for robust speaker recognition.…”

Section: Introductionmentioning

confidence: 99%

A robust polynomial regression-based voice activity detector for speaker verification

DİŞKEN

Tüfekçi

Çevik

2017

J AUDIO SPEECH MUSIC PROC.

View full text Add to dashboard Cite

Robustness against background noise is a major research area for speech-related applications such as speech recognition and speaker recognition. One of the many solutions for this problem is to detect speech-dominant regions by using a voice activity detector (VAD). In this paper, a second-order polynomial regression-based algorithm is proposed with a similar function as a VAD for text-independent speaker verification systems. The proposed method aims to separate steady noise/silence regions, steady speech regions, and speech onset/offset regions. The regression is applied independently to each filter band of a mel spectrum, which makes the algorithm fit seamlessly to the conventional extraction process of the mel-frequency cepstral coefficients (MFCCs). The kmeans algorithm is also applied to estimate average noise energy in each band for spectral subtraction. A pseudo SNR-dependent linear thresholding for the final VAD output decision is introduced based on the k-means energy centers. This thresholding considers the speech presence in each band. Conventional VADs usually neglect the deteriorative effects of the additive noise in the speech regions. Contrary to this, the proposed method decides not only for the speech presence, but also if the frame is dominated by the speech, or the noise. Performance of the proposed algorithm is compared with a continuous noise tracking method, and another VAD method in speaker verification experiments, where five different noise types at five different SNR levels were considered. The proposed algorithm showed superior verification performance both with the conventional GMM-UBM method, and the stateof-the-art i-vector method.

show abstract

Regularized All-Pole Models for Speaker Verification Under Noisy Environments

Cited by 25 publications

References 15 publications

A Review on Feature Extraction for Speaker Recognition under Degraded Conditions

A Review on Feature Extraction for Speaker Recognition under Degraded Conditions

Combining dynamic features with MFCC for text-independent speaker identification

A robust polynomial regression-based voice activity detector for speaker verification

Contact Info

Product

Resources

About