Automatic Speaker Recognition System Research Articles

Exploring the performance of automatic speaker recognition using twin speech and deep learning-based artificial neural networks.

This study assessed the influence of speaker similarity and sample length on the performance of an automatic speaker recognition (ASR) system utilizing the SpeechBrain toolkit. The dataset comprised recordings from 20 male identical twin speakers engaged in spontaneous dialogues and interviews. Performance evaluations involved comparing identical twins, all speakers in the dataset (including twin pairs), and all speakers excluding twin pairs. Speech samples, ranging from 5 to 30 s, underwent assessment based on equal error rates (EER) and Log cost-likelihood ratios (Cllr). Results highlight the substantial challenge posed by identical twins to the ASR system, leading to a decrease in overall speaker recognition accuracy. Furthermore, analyses based on longer speech samples outperformed those using shorter samples. As sample size increased, standard deviation values for both intra and inter-speaker similarity scores decreased, indicating reduced variability in estimating speaker similarity/dissimilarity levels in longer speech stretches compared to shorter ones. The study also uncovered varying degrees of likeness among identical twins, with certain pairs presenting a greater challenge for ASR systems. These outcomes align with prior research and are discussed within the context of relevant literature.

Behavioral features of the speech signal as part of improving the effectiveness of the automatic speaker recognition system

The current reality is saturated with intelligent telecommunications solutions, and automatic speaker recognition systems are an integral part of many of them. They are widely used in sectors such as banking, telecommunications and forensics. The ease of performing automatic analysis and efficient extraction of the distinctive characteristics of the human voice makes it possible to identify, verify, as well as authorize the speaker under investigation. Currently, the vast majority of solutions in the field of speaker recognition systems are based on the distinctive features resulting from the structure of the speaker's vocal tract (laryngeal sound analysis), called physical features of the voice. Despite the high efficiency of such systems - oscillating at more than 95% - their further development is already very difficult, due to the fact that the possibilities of distinctive physical features have been exhausted. Further opportunities to increase the effectiveness of ASR systems based on physical features appear after additional consideration of the behavioral features of the speech signal in the system, which is the subject of this article.

Enhancing Web Application Security: Advanced Biometric Voice Verification for Two-Factor Authentication

This paper presents a voice biometrics system implemented in a web application as part of a two-factor authentication (2FA) user login. The web-based application, via a client interface, runs registration, preprocessing, feature extraction and normalization, classification, and speaker verification procedures based on a modified Gaussian mixture model (GMM) algorithm adapted to the application requirements. The article describes in detail the internal modules of this ASR (Automatic Speaker Recognition) system. A comparison of the performance of competing ASR systems using the commercial NIST 2002 SRE voice dataset tested under the same conditions is also presented. In addition, it presents the results of the influence of the application of cepstral mean and variance normalization over a sliding window (WCMVN) and its relevance, especially for voice recordings recorded in varying acoustic tracks. The article also presents the results of the selection of a reference model representing an alternative hypothesis in the decision-making system, which significantly translates into an increase in the effectiveness of speaker verification. The final experiment presented is a test of the performance achieved in a varying acoustic environment during remote voice login to a web portal by the test group, as well as a final adjustment of the decision-making threshold.

Automatic Speaker Recognition System Based on Gaussian Mixture Models, Cepstral Analysis, and Genetic Selection of Distinctive Features

This article presents the Automatic Speaker Recognition System (ASR System), which successfully resolves problems such as identification within an open set of speakers and the verification of speakers in difficult recording conditions similar to telephone transmission conditions. The article provides complete information on the architecture of the various internal processing modules of the ASR System. The speaker recognition system proposed in the article, has been compared very closely to other competing systems, achieving improved speaker identification and verification results, on known certified voice dataset. The ASR System owes this to the dual use of genetic algorithms both in the feature selection process and in the optimization of the system's internal parameters. This was also influenced by the proprietary feature generation and corresponding classification process using Gaussian mixture models. This allowed the development of a system that makes an important contribution to the current state of the art in speaker recognition systems for telephone transmission applications with known speech coding standards.

Comparative study to realize an automatic speaker recognition system

In this research, we present an automatic speaker recognition system based on adaptive orthogonal transformations. To obtain the informative features with a minimum dimension from the input signals, we created an adaptive operator, which helped to identify the speaker’s voice in a fast and efficient manner. We test the efficiency and the performance of our method by comparing it with another approach, mel-frequency cepstral coefficients (MFCCs), which is widely used by researchers as their feature extraction method. The experimental results show the importance of creating the adaptive operator, which gives added value to the proposed approach. The performance of the system achieved 96.8% accuracy using Fourier transform as a compression method and 98.1% using Correlation as a compression method.

Automatic Speaker Recognition Using Mel-Frequency Cepstral Coefficients Through Machine Learning

Automatic speaker recognition (ASR) systems are the field of Human-machine interaction and scientists have been using feature extraction and feature matching methods to analyze and synthesize these signals. One of the most commonly used methods for feature extraction is Mel Frequency Cepstral Coefficients (MFCCs). Recent researches show that MFCCs are successful in processing the voice signal with high accuracies. MFCCs represents a sequence of voice signal-specific features. This experimental analysis is proposed to distinguish Turkish speakers by extracting the MFCCs from the speech recordings. Since the human perception of sound is not linear, after the filterbank step in the MFCC method, we converted the obtained log filterbanks into decibel (dB) features-based spectrograms without applying the Discrete Cosine Transform (DCT). A new dataset was created with converted spectrogram into a 2-D array. Several learning algorithms were implemented with a 10-fold cross-validation method to detect the speaker. The highest accuracy of 90.2% was achieved using Multi-layer Perceptron (MLP) with tanh activation function. The most important output of this study is the inclusion of human voice as a new feature set.

Nonlinear waveform distortion: Assessment and detection of clipping on speech data and systems

Speech, speaker, and language systems have traditionally relied on carefully collected speech material for training acoustic models. There is an enormous amount of freely accessible audio content. A major challenge, however, is that such data is not professionally recorded, and therefore may contain a wide diversity of background noise, nonlinear distortions, or other unknown environmental or technology-based contamination or mismatch. There is a crucial need for automatic analysis to screen such unknown datasets before acoustic model development training, or to perform input audio purity screening prior to classification. In this study, we propose a waveform based clipping detection algorithm for naturalistic audio streams and examine the impact of clipping at different severities on speech quality measurements and automatic speaker recognition systems. We use the TIMIT and NIST SRE08 corpora as case studies. The results show, as expected, that clipping introduces a nonlinear distortion into clean speech data, which reduces speech quality and performance for speaker recognition. We also investigate what degree of clipping can be present to sustain effective speech system performance. The proposed detection system, which will be released, could contribute to massive new audio collections for speech and language technology development (e.g. Google Audioset (Gemmeke et al., 2017), CRSS-UTDallas Apollo Fearless-Steps (Yu et al., 2014) (19,000 h naturalistic audio from NASA Apollo missions)).

Towards an objective comparison of feature extraction techniques for automatic speaker recognition systems

A common limitation of the previous comparative studies on speaker-features extraction techniques lies in the fact that the comparison is done independently of the used speaker modeling technique and its parameters. The aim of the present paper is twofold. Firstly, it aims to review the most significant advancements in feature extraction techniques used for automatic speaker recognition. Secondly, it seeks to evaluate and compare the currently dominant ones using an objective comparison methodology that overcomes the various limitations and drawbacks of the previous comparative studies. The results of the carried out experiments underlines the importance of the proposed comparison methodology.

Enhancement in speaker recognition for optimized speech features using GMM, SVM and 1-D CNN

Contemporary automatic speaker recognition (ASR) systems do not provide 100% accuracy making it imperative to explore different techniques to improve it. Easy access to mobile devices and advances in sensor technology, has made voice a preferred parameter for biometrics. Here, a comparative analysis of accuracies obtained in ASR with employment of classical Gaussian mixture model (GMM), support vector machine (SVM) which is the machine learning algorithm and the state of art 1-D CNN as classifiers is presented. Authors propose considering dynamic voice features along with static features as relevant speaker information in them lead to substantial improvement in the accuracy for ASR. As concatenation of features leads to the redundancy and increased computation complexity, Fisher score algorithm was employed to select the best contributing features resulting in improvement in accuracy. The results indicate that SVM and 1-D Neural network outperform GMM. Support Vector Machine (SVM), and 1-D CNN gave comparable results with 1-D CNN giving an improved accuracy of 94.77% in ASR.

Kernal based speaker specific feature extraction and its applications in iTaukei cross language speaker recognition

Extraction and classification algorithms based on kernel nonlinear features are popular in the new direction of research in machine learning. This research paper considers their practical application in the iTaukei automatic speaker recognition system (ASR) for cross-language speech recognition. Second, nonlinear speaker-specific extraction methods such as kernel principal component analysis (KPCA), kernel independent component analysis (KICA), and kernel linear discriminant analysis (KLDA) are summarized. The conversion effects on subsequent classifications were tested in conjunction with Gaussian mixture modeling (GMM) learning algorithms; in most cases, computations were found to have a beneficial effect on classification performance. Additionally, the best results were achieved by the Kernel linear discriminant analysis (KLDA) algorithm. The performance of the ASR system is evaluated for clear speech to a wide range of speech quality using ATR Japanese C language corpus and self-recorded iTaukei corpus. The ASR efficiency of KLDA, KICA, and KLDA technique for 6 sec of ATR Japanese C language corpus 99.7%, 99.6%, and 99.1% and equal error rate (EER) are 1.95%, 2.31%, and 3.41% respectively. The EER improvement of the KLDA technique-based ASR system compared with KICA and KPCA is 4.25% and 8.51% respectively.

Assessing the effects of accent-mismatched reference population databases on the performance of an automatic speaker recognition system

Automatic Speaker Recognition (ASR) systems are designed to provide the user with statistics relating to the similarity of two or more speech samples and to the typicality of those shared features in the wider population. When an ASR system is used as part of a forensic investigation, the user must decide what counts as the appropriate ‘wider population’ and select a reference database accordingly. While it has generally been held that the voices populating the reference database should be similar in accent to that of the samples under consideration, the degree to which the accents should correspond has until now not been investigated empirically. We report in this article on a study in which the composition of the reference database was systematically varied in terms of accent, using corpora of samples of Standard Southern British English and of three subvarieties spoken in North-East England (Newcastle, Sunderland, Middlesbrough).

Exploring the relationship between voice similarity estimates by listeners and by an automatic speaker recognition system incorporating phonetic features

The present study investigates relationships between voice similarity ratings made by human listeners and comparison scores produced by an automatic speaker recognition system that includes phonetic, perceptually-relevant features in its modelling. The study analyses human voice similarity ratings of pairs of speech samples from unrelated speakers from an accent-controlled database (DyViS, Standard Southern British English) and the comparison scores from an i-vector-based automatic speaker recognition system using ‘auto-phonetic’ (automatically extracted phonetic) features. The voice similarity ratings were obtained from 106 listeners who each rated the voice similarity of pairings of ten speakers on a Likert scale via an online test.Correlation analysis and Multidimensional Scaling showed a positive relationship between listeners’ judgements and the automatic comparison scores. A separate analysis of the subsets of listener responses from English and German native speaker groups showed that a positive relationship was present for both groups, but that the correlation was higher for the English listener group.This work has key implications for forensic phonetics through highlighting the potential to automate part of the process of selecting foil voices in voice parade construction for which the collection and processing of human judgements is currently needed. Further, establishing that it is possible to use automatic voice comparisons using phonetic features to select similar-sounding voices has important applications in ‘voice casting’ (finding voices that are similar to a given voice) and ‘voice banking’ (saving one's voice for future synthesis in case of an operation or degenerative disease).

High level speaker specific features modeling in automatic speaker recognition system

Spoken words convey several levels of information. At the primary level, the speech conveys words or spoken messages, but at the secondary level, the speech also reveals information about the speakers. This work is based on the high-level speaker-specific features on statistical speaker modeling techniques that express the characteristic sound of the human voice. Using Hidden Markov model (HMM), Gaussian mixture model (GMM), and Linear Discriminant Analysis (LDA) models build Automatic Speaker Recognition (ASR) system that are computational inexpensive can recognize speakers regardless of what is said. The performance of the ASR system is evaluated for clear speech to a wide range of speech quality using a standard TIMIT speech corpus. The ASR efficiency of HMM, GMM, and LDA based modeling technique are 98.8%, 99.1%, and 98.6% and Equal Error Rate (EER) is 4.5%, 4.4% and 4.55% respectively. The EER improvement of GMM modeling technique based ASR systemcompared with HMM and LDA is 4.25% and 8.51% respectively.

SELEKCJA CECH OSOBNICZYCH SYGNAŁU MOWY Z WYKORZYSTANIEM ALGORYTMÓW GENETYCZNYCH

W referacie przedstawiono system automatycznego rozpoznawania mówcy zaimplementowany w środowisku Matlab oraz pokazano sposoby realizacji i optymalizacji poszczególnych elementów tego systemu. Główny nacisk położono na wyselekcjonowanie cech dystynktywnych głosu mówcy z wykorzystaniem algorytmu genetycznego, który pozwala na uwzględnienie synergii cech podczas selekcji. Pokazano również wyniki optymalizacji wybranych elementów klasyfikatora, m.in. liczby rozkładów Gaussa użytych do zamodelowania każdego z głosów. Ponadto, podczas tworzenia modeli głosów zastosowano model głosu uniwersalnego.

Robust speaker recognition based on biologically inspired features

This paper proposes two speech parameterisation techniques for noise-robust speaker recognition: the normalised gammachirp cepstral coefficients (NGCC) and the perceptual linear predictive normalised gammachirp (PLPnGc). These techniques employ a biologically inspired auditory model that simulates the cochlea spectral behaviour. In an automatic speaker recognition (ASR) system, we consider the Gaussian mixture model-universal background model (GMM-UBM) for speaker modelling. The performances are evaluated in clean and noisy environments using Timit, Aurora, and Demand databases. The experimental results in noisy environments showed that the biologically inspired feature extraction techniques give a better recognition rate than state-of-the-art methods.

Analysis of the Effects of Supraglottal Tract Surgical Procedures in Automatic Speaker Recognition Performance

This article evaluates the impact in the performance of state-of-the-art automatic speaker recognition schemes of three surgical procedures modifying the supraglottal tract structures of speakers. To do so, a new corpus (Cuco) was recorded, containing the speech of 107 speakers before and after surgery. Speakers were divided into four groups depending on the type of surgery: tonsillectomy, functional endoscopy sinus surgery (FESS), septoplasty, and controls. The analyzed speaker recognition schemes were i-vectors, i-vectors with supervised Universal Background Model, i-vectors employing Time-delay Deep Neural Networks and x-vectors. In all cases, probabilistic linear discriminant analysis was employed in the back-end. Results show changes in the speech of patients who underwent tonsillectomy or FESS after surgery in contrast to controls or patients who had a septoplasty, where not significant variations are observed. These changes increase the Equal Error Rate (EER) of the analyzed speaker recognition schemes for the septoplasty and FESS groups when employing enrollment data recorded before the surgery. Moreover, surgery has a similar influence in the speech of female and male speakers with respect to the analyzed schemes. In consequence, results suggest that it is advisable to update the speaker's enrollment speech after three months following supraglottal tract surgery to ensure that the effects of the operation and post-operative recovery period do not influence the performance of the automatic speaker recognition systems.

A Survey of Automatic Speaker Recognition System Using Artificial Neural Networks

A Survey of Automatic Speaker Recognition System Using Artificial Neural Networks

Kalibrasi Rasio kemungkinan pada Sistem Rekognisi Pengucap Otomatis untuk Aplikasi Forensik di Indonesia

Likelihood ratio calibration is a very important step when applying an automatic speaker recognition system to forensic applications. This paper presents the process and evaluation of a speaker recognition system developed with spoken Indonesian database. The system is developed using MFCC feature, GMM-UBM modeling, and Z normalization. System performances were evaluated based on male and female genders, and two scenarios i.e., natural conversation and interview. System evaluations were done using performance measures for both discrimination and calibration abilities. Results show that based on various indicators, the system behaves very well. This is shown by the best achievable EER values of 4.66%, and Cmc values of 0.04. Therefore, the developed automatic speaker recognition system is now ready to be used for forensic applications in Indonesia.

International practices in forensic speaker comparisons

A survey relating to current practices in forensic speaker comparison testing was recently undertaken of 39 laboratories and individual practitioners across 23 countries. Questions were organised around a number of themes, including the preliminary assessment and preparation of case materials, the checking of analysts' work, frameworks used for the expression of conclusions, the use of automatic speaker recognition systems, the use of reference populations, and awareness of cognitive bias. Developmental trends in this area of forensic speech science are established by comparing responses to the present survey with those to the authors' earlier survey published in 2011.

Advantages and Disadvantages of Automatic Speaker Recognition Systems

Automatic speaker recognition systems use the machines to recognize an individual via a spoken sentence. Those systems recognize a specific individual or confirm an individual’s claimed identity. The most common type of voice biometrics is the Speaker Recognition. Its task focused on validation of a person’s claimed identity, using features that have been obtained via their voices. Throughout the last decades a wide range of new advances in the speaker recognition area have been accomplished, but there are still many problems that need solving or require enhanced solutions. In this paper, a brief overview of speech processing is given firstly, then some feature extraction and classifier techniques are described, also a comparative and analysis of some previous research are studied in depth, all this work leads to determine the best methods for speaker recognition. Adaptive MFCC and Deep Learning methods are determined to be more efficient and accurate than other methods in speaker recognition, thus these methods are recommend to be more suitable for practical applications.