New algorithm for variable-rate linear broadcast network coding

To adjust the variance of source rate in linear broadcast networks, global encoding kernels should have corresponding dimensions to instruct the decoding process. The algorithm of constructing such global encoding kernels is to adjust heterogeneous network to possible link failures. Linear algebra, graph theory and group theory are applied to construct one series of global encoding kernels which are applicable to all source rates. The effectiveness and existence of such global encoding kernels are proved. Based on 2 information flow, the algorithm of construction is explicitly given within polynomial time O（｜E｜ ｜T｜.ω^2max）, and the memory complexity of algorithm is O（｜E｜）. Both time and memory complexity of this algorithm proposed can be O（ωmax） less than those of algorithms in related works.

Adaptive bands filter bank optimized by genetic algorithm for robust speech recognition system

Perceptual auditory filter banks such as Bark-scale filter bank are widely used as front-end processing in speech recognition systems.However,the problem of the design of optimized filter banks that provide higher accuracy in recognition tasks is still open.Owing to spectral analysis in feature extraction,an adaptive bands filter bank (ABFB) is presented.The design adopts flexible bandwidths and center frequencies for the frequency responses of the filters and utilizes genetic algorithm (GA) to optimize the design parameters.The optimization process is realized by combining the front-end filter bank with the back-end recognition network in the performance evaluation loop.The deployment of ABFB together with zero-crossing peak amplitude (ZCPA) feature as a front process for radial basis function (RBF) system shows significant improvement in robustness compared with the Bark-scale filter bank.In ABFB,several sub-bands are still more concentrated toward lower frequency but their exact locations are determined by the performance rather than the perceptual criteria.For the ease of optimization,only symmetrical bands are considered here,which still provide satisfactory results.

Speech Analysis for Diagnosis of Parkinson’s Disease Using Genetic Algorithm and Support Vector Machine

Parkinson’s disease (PD) is the most common disease of motor system degeneration that occurs when the dopamine-producing cells are damaged in substantia nigra. To detect PD, various signals have been investigated, including EEG, gait and speech. Since approximately 90 percent of the people with PD suffer from speech disorders, speech analysis is considered as the most common technique for this aim. This paper proposes a new algorithm for diagnosing of Parkinson’s disease based on voice analysis. In the first step, genetic algorithm (GA) is undertaken for selecting optimized features from all extracted features. Afterwards a network based on support vector machine (SVM) is used for classification between healthy and people with Parkinson. The dataset of this research is composed of a range of biomedical voice signals from 31 people, 23 with Parkinson’s disease and 8 healthy people. The subjects were asked to pronounce letter “A” for 3 seconds. 22 linear and non-linear features were extracted from the signals that 14 features were based on F0 (fundamental frequency or pitch), jitter, shimmer and noise to harmonics ratio, which are main factors in voice signal. Because changing in these factors is noticeable for the people with PD, optimized features were selected among them. Of the various numbers of optimized features, the data classification was investigated. Results show that the classification accuracy percent of 94.50 per 4 optimized features, the accuracy percent of 93.66 per 7 optimized features and the accuracy percent of 94.22 per 9 optimized features, could be achieved. It can be observed that the best classification accuracy may be achieved using Fhi (Hz), Fho (Hz), jitter (RAP) and shimmer (APQ5).

Enhancing Parkinson’s Disease Diagnosis Accuracy Through Speech Signal Algorithm Modeling

Parkinson’s disease(PD),one of whose symptoms is dysphonia,is a prevalent neurodegenerative disease.The use of outdated diagnosis techniques,which yield inaccurate and unreliable results,continues to represent an obstacle in early-stage detection and diagnosis for clinical professionals in the medical field.To solve this issue,the study proposes using machine learning and deep learning models to analyze processed speech signals of patients’voice recordings.Datasets of these processed speech signals were obtained and experimented on by random forest and logistic regression classifiers.Results were highly successful,with 90%accuracy produced by the random forest classifier and 81.5%by the logistic regression classifier.Furthermore,a deep neural network was implemented to investigate if such variation in method could add to the findings.It proved to be effective,as the neural network yielded an accuracy of nearly 92%.Such results suggest that it is possible to accurately diagnose early-stage PD through merely testing patients’voices.This research calls for a revolutionary diagnostic approach in decision support systems,and is the first step in a market-wide implementation of healthcare software dedicated to the aid of clinicians in early diagnosis of PD.

AN EFFECTIVE LVQ-BASED ALGORITHMFOR ROBUST SPEECH RECOGNITION

Dynamic time warping (DTW) and dynamic spectral wafliing (DSW)techniques are introduced into learning vector quantization (LVQ) algorithm to con-struct a “dynamic” Bayes classifier for speech recognition. It can preduce highly dis-criminiative “dynamic” reference vectors to represent the temporal and spectral vari-abilities of speech. Recognition experiments on 19 Chinese consonants show that the“dynamic” classifier outperforms the original “static” classifier significantly.

Improved Attentive Recurrent Network for Applied Linguistics-Based Offensive Speech Detection

Applied linguistics is one of the fields in the linguistics domain and deals with the practical applications of the language studies such as speech processing,language teaching,translation and speech therapy.The ever-growing Online Social Networks(OSNs)experience a vital issue to confront,i.e.,hate speech.Amongst the OSN-oriented security problems,the usage of offensive language is the most important threat that is prevalently found across the Internet.Based on the group targeted,the offensive language varies in terms of adult content,hate speech,racism,cyberbullying,abuse,trolling and profanity.Amongst these,hate speech is the most intimidating form of using offensive language in which the targeted groups or individuals are intimidated with the intent of creating harm,social chaos or violence.Machine Learning(ML)techniques have recently been applied to recognize hate speech-related content.The current research article introduces a Grasshopper Optimization with an Attentive Recurrent Network for Offensive Speech Detection(GOARN-OSD)model for social media.The GOARNOSD technique integrates the concepts of DL and metaheuristic algorithms for detecting hate speech.In the presented GOARN-OSD technique,the primary stage involves the data pre-processing and word embedding processes.Then,this study utilizes the Attentive Recurrent Network(ARN)model for hate speech recognition and classification.At last,the Grasshopper Optimization Algorithm(GOA)is exploited as a hyperparameter optimizer to boost the performance of the hate speech recognition process.To depict the promising performance of the proposed GOARN-OSD method,a widespread experimental analysis was conducted.The comparison study outcomes demonstrate the superior performance of the proposed GOARN-OSD model over other state-of-the-art approaches.

基于两步单源点筛选的改进退化解混和估计算法

退化解混和估计(Degenerate unmixing estimation technique,DUET)算法是一种典型的欠定盲源分离算法,其采用的二进制时频掩蔽会保留部分干扰信号。提出了基于两步单源点筛选的改进DUET算法,首先使用余弦角算法进行单源点筛选,再采用计算相似度的方法进行第二步单源点筛选。通过两步单源点筛选获得更精确的目标信号和干扰信号后,设计用于抵消干扰信号的滤波器取代DUET中的二进制时频掩蔽,达到抑制干扰信号和提取目标信号的目的。仿真实验结果表明,该方法在正定盲源分离和欠定盲源分离两种情况下都有较优的盲源分离性能。

基于动量梯度下降的回声消除算法

针对极端环境话音系统下声学回波影响工作人员正常施工,且常规声学回声消除算法收敛速度慢的问题,提出一种基于动量梯度下降的基于l0范数的改进系数成比例归一化最小均方误差算法(L0⁃IPNLMS)。该算法将动量因子引入L0⁃IPNLMS算法中,解决在算法运行过程中梯度下降时梯度摆动幅度可能过大的问题,也提高了自适应滤波器的收敛速度,且残余回声下降明显,声学回波抑制效果更好。仿真实验表明,与L0⁃IPNLMS算法相比,新算法在模拟随机多音信号与真实语音信号输入时,均方误差(MSE)可以降低3.47 dB和3.69 dB,回波抑制比(ERLE)提高了3.46 dB和3.68 dB,在低信噪比情况下,使用新算法对真实语音信号进行回声消除,收敛速度高于L0⁃IPNLMS等算法,且收敛效果有明显改进。

BDO与VMD-EAM算法融合的单通道语音增强模型

为了改善VMD在单通道语音增强中关键参数选择难、声学特征缺失多的问题,该文提出了一种BDO与VMD-EAM算法融合的单通道语音增强模型。利用DBO对输入的含噪音频进行处理,得到分解模态数与惩罚因子的最佳组合,实现VMD在分解中关键参数的自适应寻优,再根据EAM的相似度评估结果完成IMF分量的分类,对噪声分量的高频信息使用小波阈值法进行滤除,最后基于IVMD-EAM模型的重构原理实现信号重构。仿真验证及实测实验表明,所构建的语音增强模型可有效抑制噪声并提升英文语音的增强效果,各项性能评价指标也显著优于其他2种单通道语音增强模型。

Annoyance-type speech emotion detection in working environment

In order to recognize people＇s annoyance emotions in the working environment and evaluate emotional well- being, emotional speech in a work environment is induced to obtain adequate samples of emotional speech, and a Mandarin database with two thousands samples is built. In searching for annoyance-type emotion features, the prosodic feature and the voice quality feature parameters of the emotional statements are extracted first. Then an improved back propagation （BP） neural network based on the shuffled frog leaping algorithm （SFLA） is proposed to recognize the emotion. The recognition capability of the BP, radical basis function （RBF） and the SFLA neural networks are compared experimentally. The results show that the recognition ratio of the SFLA neural network is 4. 7% better than that of the BP neural network and 4. 3% better than that of the RBF neural network. The experimental results demonstrate that the random initial data trained by the SFLA can optimize the connection weights and thresholds of the neural network, speed up the convergence and improve the recognition rate.

语音实验室端到端即时通信认证协议设计

为保证语音实验室端到端即时通信安全,确保语音内容不被窃听和篡改,提出一种基于混合加解密的语音实验室端到端即时通信认证协议的设计方法。利用RSA方法加解密发送方传输的即时通信会话密钥,通过3DES方法将发送方发送的明文语音信息进行加解密。在加解密过程中,加密信息打包为加密包后发送至信息接收方,信息接收方获取加密包后,使用RSA方法、3DES方法进行有效的密钥解密认证、明文语音消息解密,获取语音实验室端到端的明文语音信息。实验结果显示,此协议使用下,语音实验室端到端即时通信的认证加速比提升,且仅在密钥输入内容准确的情况下,信息接收方才可得到准确的明文语音消息内容,且不存在内容失真问题。

一种改进的广义旁瓣对消阵列语音增强算法

传统广义旁瓣对消(generalized sidelobe cancellation,GSC)算法在复杂声学环境下性能较差,语音增强效果不理想,需要加强其去噪能力以提高输出语音质量。对此,提出一种改进型广义旁瓣对消算法。构建基于频域GSC和时域GSC两级滤波的系统结构,利用一个选择滤波器输出两级滤波中质量较好的语音;并将GSC上支路权值修改为可自适应调节的形式,提高算法的适应性;GSC权值迭代时采用一种变步长自适应算法,步长因子根据信号信干噪比实时调整,防止滤波器权值发散。实验结果表明,相比于传统GSC算法,新算法在SINR、PESQ、STOI和SDR等指标上都有提升。

改进遗传算法的电子会议汉语语音识别方法

语音识别普遍存在识别不准确、不全面的问题,影响电子会议汉语记录的质量。面对这种情况,为提高语音识别性能,提出一种改进遗传算法的电子会议汉语语音识别方法。该方法通过预加重、分帧加窗以及去噪,预处理语音信号。利用改进遗传算法选取最优语音特征,语音特征包括梅尔频率倒谱系数、短时平均能量以及频谱均值。以三个特征对应数值的标准化数值为输入,利用构建的基于改进神经网络的识别模型将语音转换为对应的汉语文字,实现语音识别。结果表明,在基于改进遗传算法的识别方法应用下,误识率最高仅为2.122%,识全率最低为95.621%,由此说明所研究识别方法的识别更为准确和全面,识别效果更好。

基于MFCC提取和DTW优化的连续音频识别算法设计

介绍了一种新型的利用梅尔频率倒谱系数(MFCC)提取和动态时间规整技术(DTW)优化的连续音频识别算法。首先对数学原理与算法步骤进行设计与规划,使用大规模音频数据库进行预处理,经过时域和频域分析提取相应的特征;然后利用双门限法把连续音频切分为不同的音频块,并对切分部分进行针对性识别,将其与时频域数据库的模板进行匹配比对,实现了较好的连续音频识别效果,在时域和频域识别上的准确性均能达到89%。该研究成果可应用于钢琴教学系统的开发,尤其是在辅助学习者正确弹出曲谱方面具有广阔的应用前景。

Speech Enhancement Based on Approximate Message Passing

To overcome the limitations of conventional speech enhancement methods, such as inaccurate voice activity detector(VAD) and noise estimation, a novel speech enhancement algorithm based on the approximate message passing(AMP) is adopted. AMP exploits the difference between speech and noise sparsity to remove or mute the noise from the corrupted speech. The AMP algorithm is adopted to reconstruct the clean speech efficiently for speech enhancement. More specifically, the prior probability distribution of speech sparsity coefficient is characterized by Gaussian-model, and the hyper-parameters of the prior model are excellently learned by expectation maximization(EM) algorithm. We utilize the k-nearest neighbor(k-NN) algorithm to learn the sparsity with the fact that the speech coefficients between adjacent frames are correlated. In addition, computational simulations are used to validate the proposed algorithm, which achieves better speech enhancement performance than other four baseline methods-Wiener filtering, subspace pursuit(SP), distributed sparsity adaptive matching pursuit(DSAMP), and expectation-maximization Gaussian-model approximate message passing(EM-GAMP) under different compression ratios and a wide range of signal to noise ratios(SNRs).

An Efficient Reference Free Adaptive Learning Process for Speech Enhancement Applications

In issues like hearing impairment,speech therapy and hearing aids play a major role in reducing the impairment.Removal of noise signals from speech signals is a key task in hearing aids as well as in speech therapy.During the transmission of speech signals,several noise components contaminate the actual speech components.This paper addresses a new adaptive speech enhancement(ASE)method based on a modified version of singular spectrum analysis(MSSA).The MSSA generates a reference signal for ASE and makes the ASE is free from feeding reference component.The MSSA adopts three key steps for generating the reference from the contaminated speech only.These are decomposition,grouping and reconstruction.The generated reference is taken as a reference for variable size adaptive learning algorithms.In this work two categories of adaptive learning algorithms are used.They are step variable adaptive learning(SVAL)algorithm and time variable step size adaptive learning(TVAL).Further,sign regressor function is applied to adaptive learning algorithms to reduce the computational complexity of the proposed adaptive learning algorithms.The performance measures of the proposed schemes are calculated in terms of signal to noise ratio improvement(SNRI),excess mean square error(EMSE)and misadjustment(MSD).For cockpit noise these measures are found to be 29.2850,-27.6060 and 0.0758 dB respectively during the experiments using SVAL algorithm.By considering the reduced number of multiplications the sign regressor version of SVAL based ASE method is found to better then the counter parts.

Applying and Comparison of Chaotic-Based Permutation Algorithms for Audio Encryption

This research presents,and claries the application of two permutation algorithms,based on chaotic map systems,and applied to a le of speech signals.They are the Arnold cat map-based permutation algorithm,and the Baker’s chaotic map-based permutation algorithm.Both algorithms are implemented on the same speech signal sample.Then,both the premier and the encrypted le histograms are documented and plotted.The speech signal amplitude values with time signals of the original le are recorded and plotted against the encrypted and decrypted les.Furthermore,the original le is plotted against the encrypted le,using the spectrogram frequencies of speech signals with the signal duration.These permutation algorithms are used to shufe the positions of the speech les signals’values without any changes,to produce an encrypted speech le.A comparative analysis is introduced by using some of sundry statistical and experimental analyses for the procedures of encryption and decryption,e.g.,the time of both procedures,the encrypted audio signals histogram,the correlation coefcient between specimens in the premier and encrypted signals,a test of the Spectral Distortion(SD),and the Log-Likelihood Ratio(LLR)measures.The outcomes of the different experimental and comparative studies demonstrate that the two permutation algorithms(Baker and Arnold)are sufcient for providing an efcient and reliable voice signal encryption solution.However,the Arnold’s algorithm gives better results in most cases as compared to the results of Baker’s algorithm.

Hybrid In-Vehicle Background Noise Reduction for Robust Speech Recognition:The Possibilities of Next Generation 5G Data Networks

This pilot study focuses on employment of hybrid LMS-ICA system for in-vehicle background noise reduction.Modern vehicles are nowadays increasingly supporting voice commands,which are one of the pillars of autonomous and SMART vehicles.Robust speaker recognition for context-aware in-vehicle applications is limited to a certain extent by in-vehicle back-ground noise.This article presents the new concept of a hybrid system which is implemented as a virtual instrument.The highly modular concept of the virtual car used in combination with real recordings of various driving scenarios enables effective testing of the investigated methods of in-vehicle background noise reduction.The study also presents a unique concept of an adaptive system using intelligent clusters of distributed next generation 5G data networks,which allows the exchange of interference information and/or optimal hybrid algorithm settings between individual vehicles.On average,the unfiltered voice commands were successfully recognized in 29.34%of all scenarios,while the LMS reached up to 71.81%,and LMS-ICA hybrid improved the performance further to 73.03%.

Enhanced Marathi Speech Recognition Facilitated by Grasshopper Optimisation-Based Recurrent Neural Network

Communication is a significant part of being human and living in the world.Diverse kinds of languages and their variations are there;thus,one person can speak any language and cannot effectively communicate with one who speaks that language in a different accent.Numerous application fields such as education,mobility,smart systems,security,and health care systems utilize the speech or voice recognition models abundantly.Though,various studies are focused on the Arabic or Asian and English languages by ignoring other significant languages like Marathi that leads to the broader research motivations in regional languages.It is necessary to understand the speech recognition field,in which the major concentrated stages are feature extraction and classification.This paper emphasis developing a Speech Recognition model for the Marathi language by optimizing Recurrent Neural Network(RNN).Here,the preprocessing of the input signal is performed by smoothing and median filtering.After preprocessing the feature extraction is carried out using MFCC and Spectral features to get precise features from the input Marathi Speech corpus.The optimized RNN classifier is used for speech recognition after completing the feature extraction task,where the optimization of hidden neurons in RNN is performed by the Grasshopper Optimization Algorithm(GOA).Finally,the comparison with the conventional techniques has shown that the proposed model outperforms most competing models on a benchmark dataset.

Hidden Markov Models for Automatic Speech Recognition

In this paper the authors look into the problem of Hidden Markov Models （HMM）： the evaluation, the decoding and the learning problem. The authors have explored an approach to increase the effectiveness of HMM in the speech recognition field. Although hidden Markov modeling has significantly improved the performance of current speech-recognition systems, the general problem of completely fluent speaker-independent speech recognition is still far from being solved. For example, there is no system which is capable of reliably recognizing unconstrained conversational speech. Also, there does not exist a good way to infer the language structure from a limited corpus of spoken sentences statistically. Therefore, the authors want to provide an overview of the theory of HMM, discuss the role of statistical methods, and point out a range of theoretical and practical issues that deserve attention and are necessary to understand so as to further advance research in the field of speech recognition.