Cantonese speaker recognition system based on EM algorithm in noisy environments

<p>In Ethiopia, the largest ethnic and linguistic groups are the Oromos, Amharas and Tigrayans. This paper presents the performance analysis of text-independent speaker identification system for the Amharic language in noisy environments. VQ (Vector Quantization), GMM (Gaussian Mixture Models), BPNN (Back propagation neural network), MFCC (Mel-frequency cepstrum coefficients), GFCC (Gammatone Frequency Cepstral Coefficients), and a hybrid approach had been use as techniques for identifying speakers of Amharic language in noisy environments. For the identification process, speech signals are collected from different speakers including both sexes; for our data set, a total of 90 speakers’ speech samples were collected, and each speech have 10 seconds duration from each individual. From these speakers, 59.2%, 70.9% and 84.7% accuracy are achieved when VQ, GMM and BPNN are used on the combined feature vector of MFCC and GFCC. </p>

Traditional coal-gangue recognition methods usually do not consider the impact of equipment noise, which severely limits its adaptability and recognition accuracy. This paper mainly studies the more accurate recognition of coal-gangue in the noise site environment with the operation of shearer, conveyor, transfer machine and other device in the process of top coal caving. Mel Frequency Cepstrum Coefficients (MFCC) smoothing method was introduced to express the intrinsic feature of sound pressure more clearly in the coal-gangue recognition site. Then, a multi-branch convolution neural network (MBCNN) model with three branches was developed, and the smoothed MFCC feature was incorporated into this model to realize the recognition of falling coal and gangue in noisy environment. The sound pressure signal datasets under the operation of different device were constructed through a great deal of laboratory and site data acquisition. Comparative experiments were carried out on noiseless dataset, single noise dataset and simulated site dataset, and the results show that our method can provide higher correct recognition accuracy and better robustness. The proposed coal-gangue recognition approach based on MBCNN and MFCC smoothing can not only recognize the state of falling coal or gangue, but also recognize the operational state of site device.

In this paper a new fast motion estimation method useful for video encoding in a noisy environment is presented. Motion estimation methods can broadly categorized as Block Matching (BM) methods, Pel Recursive (PR) methods, and Frequency Domain (FD) methods. Majority of the current fast motion estimation methods belong to block matching category. These methods typically achieve speedup by either reducing the number of search locations or by reducing the number of computations at each search location. Our proposed method is based on the pel recursive formulation. However, in order to achieve fast estimation, we operate on a block of pixels. To provide a robust behavior in a noisy environment we use a Total Least Squares (TLS) based estimation scheme which aims at estimating the true motion vectors. The simulation results indicate the speedup and robustness of the algorithm in noisy environment.

Fibonacci multi-modal optimization algorithm in noisy environment

In recent years, crime prevention systems have been developed to detect various hazardous situations. In general, the systems utilize the image information recorded by a camera to monitor the situations. It is however difficult to detect them in the blind area. To address the problem, it is required to utilize not only image information but also acoustic information occurred in such situations. Our previous study showed that two acoustic features including rahmonic and mel-frequency cepstrum coefficients (MFCCs) are effective for detecting the shouted speech. Rahmonic shows a subharmonic of fundamental frequency in the cepstrum domain, and MFCCs represent coefficients that collectively make up mel-frequency cepstrum. In this method, a shouted speech model is constructed from these features by using a gaussian mixture model (GMM). However, the previous method with GMM has difficulty in representing temporal changes of the speech features. In this study, we further expand the previous method using hidden Markov model (HMM) which has state transition to represent the temporal changes. Through objective experiments, the proposed method using HMM could achieve higher detection performance of the shouted speech than the conventional method using GMM.

The model-based speech enhancement method usually models types of noises in a prior and selects one noise model in the enhancement phase. In this paper, we study the modeling and selection (i.e. noise identification) algorithm for the model-base speech enhancement. For the noise signal features, Mel Frequency Cepstrum Coefficient (MFCC), Liner Prediction Coefficient (LPC), Linear Spectral Frequency (LSF) and Linear Predictive Cepstrum Coefficients (LPCC) are utilized. For the noise model, codebook, Gaussian Mixture Model (GMM) and Hidden Markov Model (HMM) are adopted. We couple these features and models and provide a comparative performance analysis of them. In the experiment, we use NOIZEUS database for noise modeling and testing. The experiment results show that MFCC and GMM is an excellent couple for noise identification, and LFS can take the place of LPC in the codebook-based speech enhancement.

The reverberation power spectral density (PSD) is often required for dereverberation and noise reduction algorithms. In this work, we compare two maximum likelihood (ML) estimators of the reverberation PSD in a noisy environment. In the first estimator, the direct path is first blocked. Then, the ML criterion for estimating the reverberation PSD is stated according to the probability density function of the blocking matrix (BM) outputs. In the second estimator, the speech component is not blocked. Instead, the ML criterion for estimating the speech and reverberation PSD is stated according to the probability density function of the microphone signals. To compare the expected mean square error (MSE) between the two ML estimators of the reverberation PSD, the Cramer–Rao Bounds (CRBs) for the two ML estimators are derived. We show that the CRB for the joint reverberation and speech PSD estimator is lower than the CRB for estimating the reverberation PSD from the BM outputs. Experimental results show that the MSE of the two estimators indeed obeys the CRB curves. Experimental results of multimicrophone dereverberation and noise reduction algorithm show the benefits of using the ML estimators in comparison with another baseline estimators.

Genetic algorithms are adaptive search techniques which have been used to learn high-performance knowledge structures in reactive environments that provide information in the form of payoff. In general, payoff can be viewed as a noisy function of the structure being evaluated, and the learning task can be viewed as an optimization problem in a noisy environment. Previous studies have shown that genetic algorithms can perform effectively in the presence of noise. This work explores in detail the tradeoffs between the amount of effort spent on evaluating each structure and the number of structures evaluated during a given iteration of the genetic algorithm. Theoretical analysis shows that, in some cases, more efficient search results from less accurate evaluations. Further evidence is provided by a case study in which genetic algorithms are used to obtain good registrations of digital images.

Gaussian mixture model based parameter estimation and classification has recently received great attention in modelling and processin g data. Gaussian Mixture Model (GMM) is the probabilistic model for representing the presence of subpopulations and it works well with the classification and parameter estimation strategy. Here in this work Maximum Likelihood Estimation (MLE) based on Expectation Maximization (EM) is being used for the parameter estimation approach and the estimated parameters are being used for the training and the testing of the images for their normality and the abnormality. With the mean and the covariance calculated as the parameters they are used in the Gaussian Mixture Model (GMM) based training of the classifier. Support Vector Machine a discriminative classifier and the Gaussian Mixture Model a generative model classifier are the two most popular techniques. The performance of the classification strategy of both the classifiers used has a better proficiency when compared to the other classifiers. By combining the SVM and GMM we co uld be able to classify at a better level since estimating the parameters through the GMM has a very few amount of features and hence it is not needed to use any of the feature reduction techniques. In this the GMM classifier and the SVM classifier are trained usin g the parameters and they are to be compared.

The traditional Expectation-Maximization (EM) training of Gaussian Mixture Model (GMM) is essentially a batch mode procedure which requires the multiple data samples with the sufficient size to update the model parameters. This severely limits the deployment and adaptation of GMM in many real-time online systems since the newly observed data samples are expected to be incorporated into the system upon available via retraining the model. This paper presents a new online incremental EM training procedure of GMM, which aims to perform the EM training incrementally and so can adapt GMM online sample by sample. The proposed method is extended on two kinds of EM algorithms for GMM, namely, Split-and-Merge EM and the traditional EM. Experiments on both the synthetic data and a speech processing task show the advantages and efficiency of the new method.

Automatic Speech Emotion Recognition (SER) is a current research topic in the field of Human Computer Interaction (HCI) with a wide range of applications. The purpose of speech emotion recognition system is to automatically classify speaker's utterances into different emotional states such as disgust, boredom, sadness, neutral, and happiness. The speech samples in this paper are from the Berlin emotional database. Mel Frequency cepstrum coefficients (MFCC), Linear prediction coefficients (LPC), linear prediction cepstrum coefficients (LPCC), Perceptual Linear Prediction (PLP) and Relative Spectral Perceptual Linear Prediction (Rasta-PLP) features are used to characterize the emotional utterances using a combination between Gaussian mixture models (GMM) and Support Vector Machines (SVM) based on the Kullback-Leibler Divergence Kernel. In this study, the effect of feature type and its dimension are comparatively investigated. The best results are obtained with 12-coefficient MFCC. Utilizing the proposed features a recognition rate of 84% has been achieved which is close to the performance of humans on this database.

In environments characterized by elevated noise levels, such as airports or aircraft cabins, travelers often find themselves involuntarily speaking loudly and drawing closer to one another in an effort to enhance communication and speech intelligibility. Unfortunately, this unintentional behaviour increases the risk of respiratory particles dispersion, potentially carrying infectious agents like bacteria which makes the contagion control more challenging. The accurate characterization of the risk associated to speaking, in such a challenging noise environment with multiple overlapping speech sources, is therefore of outmost importance. Among the most advanced signal processing strategies that can be used to accurately determine who spoke when and with whom and for how long but most importantly how loudly, at one location, artificial intelligence-based speaker diarization approaches were considered and adapted for this task. This article details the implementation of speaker diarization algorithms, customized to extract speaker and speech parameters discreetly. Validation and preliminary study results are also provided. The algorithms calculate speech duration and sound pressure level for each sentence and speaker, aiding in assessing viral contaminant spread. The paper focuses on applying these algorithms in noisy environments, particularly in confined spaces with multiple speakers. The findings contribute to proactive measures for containing and managing communicable diseases in air travel settings.

This paper analyses the effect of noise on the performance of a changing range genetic algorithm (CRGA). CRGA adaptively shifts and shrinks the size of the search space of the feasible region by employing feasible and infeasible solutions in the population to reach the global optimum. An additional modification of CRGA was introduced to reduce the effects of noise on the performance of the algorithm. Several test cases demonstrated the ability of the improved CRGA to deal with an additive noise. Copyright © 2007 John Wiley & Sons, Ltd.

This study tested a combination of algorithms designed to improve cochlear implant performance in noise. A noise reduction (NR) algorithm, based on signal to noise ratio estimation was evaluated in combination with several directional microphone algorithms available in the Cochlear CP810 sound processor. Fourteen adult unilateral cochlear implant users participated in the study. Evaluation was conducted using word recognition in quiet, sentence recognition in noise, and subjective feedback via questionnaire after a period of take-home use. Music appreciation was also evaluated in a controlled listening task. The sentence recognition task measured speech reception threshold for 50% morphemes correct. The interfering maskers were speech-weighted noise and competing talkers, which were spatially separated from the target speech. In addition, the locations of the noise maskers changed during the test in an effort to replicate relevant real-world listening conditions. SmartSound directionality settings Standard, Zoom, and Beam (used in the SmartSound programs Everyday, Noise, and Focus, respectively) were all evaluated with and without NR. Microphone directionality demonstrated a consistent benefit in sentence recognition in all noise conditions tested. The group average speech reception threshold benefit over the Standard setting was 3.7 dB for Zoom and 5.3 dB for Beam. Addition of the NR algorithm further improved sentence recognition by 1.3 dB when the noise maskers were speech-weighted noise. There was an overall group preference for the NR algorithm in noisy environments. Group mean word recognition in quiet, preference in quiet conditions, and music appreciation were all unaffected by the NR algorithm. Multimicrophone directionality was effective in improving speech understanding in spatially separated noisy conditions. The single-channel NR algorithm further enhanced speech intelligibility in speech-weighted noise for cochlear implant users while maintaining equivalent performance in quiet situations and when listening to music.

Objective. Muscle artifacts (MA) and electrode motion artifacts (EMA) in electrocardiogram (ECG) signals lead to a large number of false alarms from cardiac diagnostic systems. To reduce false alarms, it is necessary to improve the performance of the diagnostic algorithm in noisy environments by removing excessively noisy signals. However, existing methods focus on signal quality assessment and contain too many artificial features. Here, we present a novel method to flexibly eliminate noisy signals without any artificial features. Approach. Our method contains an improved lightweight deep neural network (DNN) to capture the signal portions damaged by EMA and MA, uses the sample entropy to quantize noisy portions, and discards those portions that exceed a defined threshold. Our method was tested in conjunction with Pan-Tompkins (PT), Filter Bank (FB), and ‘UNSW’ R-peak detection algorithms along with two heartbeat classification algorithms on datasets synthesized from the MIT-BIH Noise Stress Test Database, the China Physiological Signal Challenge 2018 Database and the MIT-BIH Arrhythmia Database. Main results. For PT R-peak detection algorithms, the sensitivity (Se) increased noticeably from 89.01% to 99.42% in the synthesized datasets with a signal-to-noise ratio of 6 dB. With the same datasets, the Se of the FB algorithm increased about 9.29%, and a 3.64% increase occurred in the Se of the ‘UNSW’ algorithm. For heartbeat classification algorithms, the overall F1-score increased about 6% in the synthesized one-heartbeat datasets. It is the first study to utilize a DNN to capture noisy segments of the ECG signal. Significance. Too many false alarms can cause alarm fatigue. Our method can be utilized as the preprocessing before signal analysis, thereby reducing false alarms from the ECG diagnostic systems.