Time encoded signal processing and recognition with vector quantization: applied to Arabic numerals

Speaker identification may be employed as part of a security system requiring user authentication. In this case, the claimed identity of the user is known from a magnetic card and PIN number, for example, and an utterance is requested to confirm the identity of the user. A fast response is necessary in the confirmation phase and a fast registration process for new users is desirable. The time encoded signal processing and recognition (TESPAR) digital language is used to preprocess the speech signal. A speaker cannot be identified directly from the single TESPAR vector since there is a highly nonlinear relationship between the vector's components such that vectors are not linearly separable. Therefore the vector and its characteristics suggest that classification using a neural network will provide an effective solution. Good classification performance has been achieved using a probabilistic RAM (pRAM) neuron. Four probabilistic pRAM neural network architectures are presented. A performance of approximately 97% correct classifications has been obtained, which is similar to results obtained elsewhere (M. Sharma and R.J. Mammone, 1996), and slightly better than a MLP network. No speech recognition stage was used in obtaining these results, so the performance relates only to identifying a speaker's voice and is therefore independent of the spoken phrase. This has been achieved in a hardware-realizable system which may be incorporated into a smart-card or similar application.

Rolling bearing performance degradation assessment has been receiving much attention for which itscrucial role to realize CBM(condition-based maintenance).This paper proposed a novel bearing performance degradation method based on TESPAR(Time Encoded Signal Processing and Recognition)and GMM(Gauss Mixture Model). TESPAR is used to extracted features which constitute A-matrix. GMM is utilized to approximate the density distribution of singular values decomposed by A-matrix. TENLLP(Time-Encoded Negative Log Likelihood Probability) serves as a fault severity which can display the similarity of the singular values between normal samples and fault samples as quantificational. Results of its application to bearing fatigue test show that this performance degradation assessment can detect the incipient rolling bearing fault and be sensitive to the change of fault.

This is the author’s version of a work that was submitted/accepted for pub-lication in the following source:

This paper proposes an automatic stream-weight optimization method for noise-robust speaker verification using multi-stream HMMs integrating spectral and prosodic information. The paper first shows the effectiveness of the multi-stream technique in our speaker verification framework. Next, a stream-weight adaptation method combining the linear discriminant analysis (LDA) and Adaboost techniques is proposed. Experiments were conducted using four-connected-digit utterances of Japanese contaminated by white noise with various SNRs. Experimental results show that 1) the verification performance was improved in all SNR conditions by using stream weights estimated by the LDA and 2) the performance is further improved by using the Adaboost in 10 - 30dB SNR conditions.

This paper proposes an automatic stream-weight and threshold estimation method for noise-robust speaker verification using multi-stream HMMs integrating segmental and prosodic information. The proposed method simultaneously optimizes stream-weights and a decision threshold by combining the linear discriminant analysis (LDA) and Adaboost techniques. Experiments were conducted using Japanese connected digit speech contaminated by white noise with various SNRs. In this experiment, a target ratio of false acceptance rate (FAR) and false rejection rate (FRR) was set by 1:1 so as to adjust them to approach an equal error rate (EER). Experimental results show that the proposed method effectively estimates stream-weights and thresholds so that FARs and FRRs are adjusted to EERs in most of the SNR conditions

The paper presents an economic solution that may be used in speaker authentication applications or in biomedical waveform diagnostics (EEG, ECG). The usual approaches require important computational resources generally employing complex signal processing methods in the frequency domain. This work proposes a low-complexity/low-cost implementation of the TESPAR (Time Encoded Signal Processing and Recognition) coding method on an ATMEL microcontroller platform. The computational requirements for this method are two orders of magnitude less than that required by other usual methods. The performed experiments studied the variation of the TESPAR-S matrices as a function of the password utterance manner: exclaiming, interrogative, slow, fast and normal and as function of the password type: a common password for all enrolled speakers and different passwords. The best results were provided in the case of cooperating utterances, when the speed of uttering the password was normal.

With the increasing number and popularity of smart devices in the past few decades, especially those that execute different types of health-related applications, their security is a growing concern. Biometric authentication can be used for device security, and speaker recognition (SR) is one of its elegant forms. In this study, an SR system is developed for the security of smart devices, which identifies a person from the shortest utterance of Arabic language, achieving the highest SR rate, by using the benchmark Linguistic Data Consortium (LDC) Arabic dataset. This study focuses on the use of the consonants and vowels of the Arabic language. It is observed that certain consonants or a fusion of certain consonants in Arabic language helps to ascertain the speaker’s identity efficiently. The shortest utterance that better presents the speaker’s identity by using such consonants is revealed. The system was analyzed by using different numbers of consonants and achieved the maximum (100%) SR rate using a fusion of only three consonants of the Arabic language. These consonants or their combinations can be used to develop text for any speaker-based authentication system.

In this paper, we propose a multi-modal biometrics system based on the face and signature recognition. For this, we suggest biometric algorithms for the face and signature recognition. First, we describe a fuzzy linear discriminant analysis (LDA) method for the face recognition. It is an expanded version of the Fisherface method using the fuzzy logic which assigns fuzzy membership to the LDA feature values. On the other hand, the signature recognition has the problem that its performance is often deteriorated by signature variation from various factors. Therefore, we propose a robust online signature recognition method using LDA and so-called Partition Peak Points (PPP) matching technique. Finally, we propose a fusion method for multi-modal biometrics based on the support vector machine. From the various experiments, we find that the proposed method renders higher recognition rates comparing with the single biometric cases under various situations.KeywordsSupport Vector MachineFeature VectorFace RecognitionLinear Discriminant AnalysisFace ImageThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

In this paper the Linear Discriminant Analysis (LDA) techniques is used to discriminate various orange juice samples [1] of pure juice, pure juice adulterated with sugar, tap water etc. Orange juices are a popular refreshment beverage and nutrition source. However, the adulteration of pure juice has now become a common practice. Adulteration ranges in sophistication from simply diluting juice with water to adding sugar, the adulterant that is most difficult to detect. Introducing these ingredients is not illegal; however, knowingly selling the resulting product as pure juice constitutes fraud. Here, the sample orange juice samples (pulpy orange) has been taken from coca cola brand. By forming samples phase data into matrix form at different frequencies in Constant phase (property of electrode-electrolyte interface of Polymer coated Copper probe dipped in ionic solution) region as well as data taken at different time. Euclidean distance from any unknown sample with the given known samples has been calculated using LDA algorithm. The sample with shortest distance will show the best maximal characteristics of the unknown sample. The MATLAB software has been used for the whole calculation and discrimination plotting of data.

In the biometrics community, face and speaker recognition are mature fields in which several systems have been proposed over the past twenty years. While existing systems perform well under controlled recording conditions, mismatch caused by the use of different sensors or a lack of cooperation from the subject still significantly affects performance, especially in challenging scenarios such as in forensics. Furthermore, existing methods suffer from scalability issues, which prevents them from taking advantage of increasingly large amounts of training data. This is otherwise a promising approach to improve accuracy in such challenging scenarios. In this thesis we address these problems of mismatch and complexity by developing scalable probabilistic models that we apply to face, speaker and bimodal recognition. Our contributions are four-fold. First, we propose a unified framework for session variability modeling techniques based on Gaussian mixture models (GMM), that encompasses inter-session variability (ISV) modeling, joint factor analysis (JFA) and total variability (TV) modeling. Second, we propose a novel exact and scalable formulation of probabilistic linear discriminant analysis (PLDA), which is a probabilistic and generative framework that models between-class and within-class variations. This formulation solves a major scalability issue, by improving both the time complexity of the training procedure from cubic to linear with respect to the number of samples per class, and the complexity of the scoring procedure. Furthermore, the implementations of all the proposed techniques are integrated into a novel collaborative open source software library called Bob 1 that enforces fair evaluations and encourages reproducible research. Fourth and finally, large-scale experiments are conducted with all of the above machine learning algorithms on several databases such as FRGC for face recognition, NIST SRE12 for speaker recognition and MOBIO for bimodal recognition, showing competitive performances.

A novel approach for water quality classification based on the integration of deep learning and feature extraction techniques

In the i‐vector/probabilistic linear discriminant analysis (PLDA) technique, the PLDA backend classifier is modelled on i‐vectors. PLDA defines an i‐vector subspace that compensates the unwanted variability and helps to discriminate among speaker‐phrase pairs. The channel or session variability manifested in i‐vectors are known to be nonlinear in nature. PLDA training, however, assumes the variability to be linearly separable, thereby causing loss of important discriminating information. Besides, the i‐vector estimation, itself, is known to be poor in case of short utterances. This paper attempts to address these issues using a simple hierarchy‐based system. A modified fuzzy‐clustering technique is employed to divide the feature space into more characteristic feature subspaces using vocal source features. Thereafter, a separate i‐vector/PLDA model is trained for each of the subspaces. The sparser alignment owing to subspace‐specific universal background model and the relatively reduced dimensions of variability in individual subspaces help to train more effective i‐vector/PLDA models. Also, vocal source features are complementary to mel frequency cepstral coefficients, which are transformed into i‐vectors using mixture model technique. As a consequence, vocal source features and i‐vectors tend to have complementary information. Thus using vocal source features for classification in a hierarchy tree may help to differentiate some of the speaker‐phrase classes, which otherwise are not easily discriminable based on i‐vectors. The proposed technique has been validated on Part 1 of RSR2015 database, and it shows a relative equal error rate reduction of up to 37.41% with respect to the baseline i‐vector/PLDA system.

We briefly review the current trend in handling the neonatal baby weight, gestational age, and other related categorical and continuous variable data, through both robust and outlier-based methods. Graphs are generally presented as smooth curves which give a wrong visual impression and totally hide the actual variability of the data, robust methods are applied without considering that they may certainly be affected by asymmetrical contamination, and the outlier-based methods are incorrectly applied without ascertaining if the individual data arrays are normally distributed, free from discordant observations. Aim: To demonstrate our new procedure for handling of neonatal baby data from two case studies. Methodology: The univariate data were handled through the application of discordancy tests in the light of new precise and accurate critical values, and significance tests were applied to both original and discordant outlier-free data arrays. The bivariate data were also handled from identifying and separating discordant outliers and reporting the quality of the coefficients in terms of standard errors. Finally, the supervised multivariate technique of linear discriminant and canonical analysis is successfully applied. Results: The results of significance tests for some cases varied from the use of all data versus censored data. The identification and separation of bivariate discordant outliers drastically improved the quality of regression lines. Linear discriminant and canonical analysis provided high success values for laboratory-controlled experiments. Conclusion: The need of obtaining both central tendency and dispersion parameters in all medical data is established. The reporting of regression coefficients with individual errors is recommended.

A modified linear discriminant analysis technique for speaker verification, referred to as normalized discriminant analysis (NDA), is presented. Using this technique it is possible to design an efficient linear classifier with very limited training data and to generate normalized discriminant scores with comparable magnitudes for different classifiers. The NDA technique is applied to a classifier for speaker verification based on speaker specific information obtained when utterances are processed with speaker independent models. In experiments conducted on a network based telephone database, the NDA technique provides an equal-error rate of 6.13% while the classifier using Fisher linear discriminant analysis has an equal-error rate of 18.18%. Furthermore, when the NDA combined with HMM approach in a hybrid speaker verification system, the rate was reduced from 5.30% (HMM with cohort normalization) to 4.32%.

ABSTRACT: Speech is one of the most natural and efficient forms of communication among humans, and it is becoming increasingly integral to human-computer interaction in modern applications such as virtual assistants, smart devices, and automated customer service systems. Among the various technologies that leverage speech, speaker recognition stands out as a biometric method that identifies or verifies individuals based on the unique characteristics of their voice. This approach has gained significant traction due to its non-intrusive nature, ease of integration, and applicability across various domains such as security, forensics, and personalised user experiences. In this paper, we introduce a novel speaker recognition system specifically designed to identify speakers based on utterances in the Marathi language, a linguistically rich and widely spoken regional language in India. The system utilises Mel-Frequency Cepstral Coefficients (MFCCS) to extract distinguishing vocal features that closely mimic the human auditory perception. MFCCS are particularly effective in capturing speech’s phonetic and acoustic properties, making them a preferred choice in speech and speaker recognition tasks. Vector quantisation (VQ) is applied to optimise the feature set and reduce computational complexity. VQ compresses the high-dimensional MFCC feature vectors into representative clusters without significantly compromising accuracy, thereby enhancing system efficiency. These features are then processed using hidden Markov Models (HMMS), which are adept at modelling temporal sequences and dynamic variations in speech. HMMS offer a statistical framework that effectively captures the sequential nature of speech patterns, leading to more reliable speaker modelling and recognition.