The Mel-frequency cepstral coefficient (MFCC) is the most widely used feature in speech and speaker recognition. However, MFCC is very sensitive to noise interference, which tends to drastically degrade the performance of recognition systems because of the mismatches between training and testing. In this paper, the logarithmic transformation in the standard MFCC analysis is replaced by a combined function to improve the noisy sensitivity. The proposed feature extraction process is also combined with speech enhancement methods, such as spectral subtraction and median-filter to further suppress the noise. Experiments show that the proposed robust MFCC-based feature significantly reduces the recognition error rate over a wide signal-to-noise ratio range.

Improved MFCC-Based Feature for Robust Speaker Identification

One trend of current research activity in speaker verification systems focuses on features that capture complementary information to the Mel-frequency cepstral coefficients (MFCCs). In conventional speaker verification methods, MFCCs are usually computed from a single-tapered DFT spectrum. These methods use the power of the Fourier transform of the time-domain speech frames and ignore the phase component due to its large amount of uncertainty. The multitaper phase information has been also ignored when the multitaper MFCC method applies. We propose a phase information extraction method that normalizes the change variation in the multitaper phase. The goal of this work is to incorporate phase information using low-variance multitaper spectrum estimation method instead of conventional single-taper window and normalizes the change variation in multitaper phase according to the frame position of the input speech to reduce the uncertainty of multitaper phase information in both the state-of-the-art Gaussian mixture model-universal background model (GMM-UBM) baseline and the i-vector speaker verification system. The experiments are conducted on the TIMIT database for both the clean and noisy conditions (SNR = 10 dB). The multitaper inverted-MFCC is also added to the overall system to produce diversity with respect to the fused subsystem. The results show relative improvement of 34% and 25% in terms of equal error rate (EER) over the MFCC baseline in clean and noisy conditions for GMM-UBM system, respectively. The relative improvement in terms of EER with i-vector back-end are about 36% and 16% in clean and noisy conditions as well.

/pdf/multitaper-mfcc-and-normalized-multitaper-phase-based-3wtswuznhi.pdf

Multitaper MFCC and normalized multitaper phase-based features for speaker verification

In speech processing for speaker verification, feature subset selection is one of the key components. Feature Subset Selection (FS) also played a vital role in the fields like pattern recognition, image processing, data mining, and gene selection. In a real world problem related to speech domain, speech sample contains a large number of relevant and irrelevant features. To increase the speaker verification rate, one needs to use the optimization technique for feature selection after the feature extraction technique. The ultimate goal is to select the most relevant subset of features for error free optimized classification in the speech domain. In this regard a novel feature subset selection algorithm is proposed using Bat algorithm and Multi Objective Optimization technique. Results of the experiment shows the proposed algorithm surpassed the accuracy rates shown by the conventional systems.

Speaker verification with optimized feature subset using MOBA

Speaker recognition is the process of recognizing a speaker from his speech. This can be used in many aspects of life, such as taking access remotely to a personal device, securing access to voice control, and doing a forensic investigation. In speaker recognition, extracting features from the speech is the most critical process. The features are used to represent the speech as unique features to distinguish speech samples from one another. In this research, we proposed the use of a combination of Wavelet and Mel Frequency Cepstral Coefficient (MFCC), Wavelet-MFCC, as feature extraction methods, and Hidden Markov Model (HMM) as classification. The speech signal is first extracted using Wavelet into one level of decomposition, then only the sub-band detail coefficient is used as the feature for further extraction using MFCC. The modeled system was applied in 300 speech datasets of 30 speakers uttering “HADIR” in the Indonesian language. K-fold cross-validation is implemented with five folds. As much as 80% of the data were trained for each fold, while the rest was used as testing data. Based on the testing, the system's accuracy using the combination of Wavelet-MFCC obtained is 96.67%.
ABSTRAK: Pengecaman penutur adalah proses mengenali penutur dari ucapannya yang dapat digunakan dalam banyak aspek kehidupan, seperti mengambil akses dari jauh ke peranti peribadi, mendapat kawalan ke atas akses suara, dan melakukan penyelidikan forensik. Ciri-ciri khas dari ucapan merupakan proses paling kritikal dalam pengecaman penutur. Ciri-ciri ini digunakan bagi mengenali ciri unik yang terdapat pada sesebuah ucapan dalam membezakan satu sama lain. Penyelidikan ini mencadangkan penggunaan kombinasi Wavelet dan Mel Frekuensi Pekali Cepstral (MFCC), Wavelet-MFCC, sebagai kaedah ekstrak ciri-ciri penutur, dan Model Markov Tersembunyi (HMM) sebagai pengelasan. Isyarat penuturan pada awalnya diekstrak menggunakan Wavelet menjadi satu tahap penguraian, kemudian hanya pekali perincian sub-jalur digunakan bagi pengekstrakan ciri-ciri berikutnya menggunakan MFCC. Model ini diterapkan kepada 300 kumpulan data ucapan daripada 30 penutur yang mengucapkan kata "HADIR" dalam bahasa Indonesia. Pengesahan silang K-lipat dilaksanakan dengan 5 lipatan. Sebanyak 80% data telah dilatih bagi setiap lipatan, sementara selebihnya digunakan sebagai data ujian. Berdasarkan ujian ini, ketepatan sistem yang menggunakan kombinasi Wavelet-MFCC memperolehi 96.67%.

/pdf/wavelet-detail-coefficient-as-a-novel-wavelet-mfcc-features-2uwazj0t.pdf

Wavelet detail coefficient as a novel wavelet-mfcc features in text-dependent speaker recognition system

En este proyecto se desarrolla un sistema de reconocimiento de locutor texto dependiente con base a los coeficientes cepstrales de la voz utilizando una Raspberry Pi 3 B, en conjunto con una tarjeta de audio USB y un microfono con conexion tipo jack de 3.5mm. 
En este dispositivo embebido se implementa la captura de la senal de audio, el acondicionamiento de la senal, la extraccion de las caracteristicas de la voz, la tecnica de reconocimiento de locutor, el manejo de la base de datos para cada los usuarios del sistema, el registro de acceso de los usuarios y la implementacion de la interfaz. 
Todo el desarrollo del sistema se realiza en software libre utilizando el lenguaje de programacion Python en la version 3.5, utilizando la libreria pyaudio para la captura de la senal, librerias matematicas como numpy, scipy, wave para el acondicionamiento de la senal, para la extraccion de los parametros cepstrales de la voz se utiliza la libreria speech_features, la tecnica de reconocimiento implementada es DTW y la base de datos del sistema es implementada en MySQL.
Al final, se realizan pruebas para identificar falsas aceptaciones y falsos rechazos y asi, determinar el porcentaje de error del sistema, su confiabilidad y eficiencia.

Sistema de identificación de locutor texto dependiente en Raspberry PI 3 B con aplicación en control de acceso.

In this paper an efficient approach for the recognition of a speaker based on text dependent speech is presented. Speaker Recognition/ Verification system suffers with wide variety of problems. In the proposed approach, the features are extracted using two methods such as Mel Frequency Cepstral Coefficients and wavelet subband coefficients, and then these futures are fused through concatenation to give optimum performance. Those concatenated feature set are more reliable to discriminate an imposter from the genuine. Those concatenated features are classified using support vector machine classifier. Performance of the proposed approach is validated on a self generated corpus of size 300 samples of 20 individual. The proposed method outperforms other existing methods.

Syed. Sharrefaunnisa

Papers

An efficient text dependent speaker recognition using fusion of MFCC and SBC