基于概率神经网络的数字音识别

本文主要利用概率神经网络和动态时间规整技术来实现数字音的识别研究。结论是在利用概率神经网络进行语音识别时可以达到比较高的识别率,此外动态时间规整函数的加入,解决了神经网络的模板规整问题。作为语音识别技术的基础,其中包含了小波的基础理论,语音的预处理,DTW技术,端点检测等基础技术。对于神经网络的加入,更加有利于深入了解神经网络这一新兴技术。     

基于机器学习的语音情感识别技术研究

针对快速发展的语音情感识别技术,归纳总结了机器学习算法在语音情感识别领域的发展过程并预测语音情感识别技术的发展方向.首先针对语音情感的离散描述模型,总结语音情感识别模型训练和识别算法的一般过程;然后,根据机器学习算法的发展阶段,分别对比分析传统机器学习算法,深度神经网络、卷积神经网络、循环神经网络等深度学习算法与语音情...     

一种基于自组织神经网络的语音识别系统

建立了一种基于自组织神经网络的语音识别系统。对语音信号进行了预处理,提取了语音信号的线性预测系数、线性预测倒谱系数和Mel倒谱特征系数,建立了基于自组织神经网络的识别判决模型。深入分析和改进了自组织神经网络的分类聚类能力,通过加强训练和设定阈值函数的方法,有效地确定了边界神经元的归属,划分出了合理的输出模式类。验证了自组织神经网络适合于处理孤立词语音识别,并具有快速性和结构简单等特征。MATLAB仿真实验表明,语音识别率达到96%。     

基于CTC-RNN的语音情感识别方法

语音情感识别的主要目的是对语音信号按照不同的情感进行分类,比如生气、恐惧、厌恶、高兴等,探究语音情感识别的任务,使用的方法是在小的语音区间上计算的一系列声学特征训练的深度递归神经网络。同时,使用CTC损失函数考虑到了同时包含情绪化和中性成分的长话语,在IEMOCAAP语料库上设置对照实验,验证了该方法的高性能。     

基于HMM和LVQ网络混合模型的语音识别方法

提出一种基于隐马尔可夫模型(HMM)和学习向量量化(LVQ)神经网络的语音识别方法.该方法先用HMM生成最佳语音状态序列,然后用函数逼近技术产生对最佳状态序列进行时闻归正,最后通过LVQ神经网络进行分类识别.理论和实验结果表明,混合模型的识别率明显高于隐马尔可夫模型的识别率.     

有序聚类方法及其在神经网络语音识别中的应用

本文提出了一种新的网络结构,我们称之为有序聚类网络。这种网络能够对语音信号进行特征提取,很好地解决神经网络语音识别中的时间规整问题。有序聚类网络从输入语音信号的特征矢量序列中撮出一组固定数目的特 矢量,然后将这组特征矢量馈入神经网络分类器进行识别。和其他的神经网络语音识别方法相比较,用这种网络进行前端处理,可以缩短后端神经网络分类器的训练和识别时间,简化经分类器的网络产高的识别率。根据该 们建立了     

数字图像识别的代价函数选择和性能评价

针对传统二次代价函数在卷积神经网络训练过程中图像识别准确率不高的问题,提出基于交叉熵代价函数的卷积神经网络算法。经数学推导,证明交叉熵代价函数较二次代价函数在图像识别中精度更高,并应用MNIST数据集和CIFAR-10数据集,使用AlexNet卷积神经网络,分别采用二次代价函数和交叉熵代价函数对图像识别模型进行训练,当数字图像识别精确率和损失值稳定后,使用测试数据对代价函数进行多次测试,对比识别准确率。仿真结果表明,此方法不仅能提高数字图像识别的准确率,而且相较于传统的代价函数,训练模型速度更快,明显缩减了训练深度神经网络模型的过程。     

基于神经网络的语音信号识别与分类

针对基于深度神经网络的端对端的语音识别技术展开研究,通过深度全序列卷积神经网络（DFCNN）声学模型和Transformer语言模型搭建一种端对端的语音识别系统。该系统完成对模型及数据的训练,实现对多字符中文语音的识别,并对隐马尔可夫语音识别方法和深度神经网络下的语音识别方法的系统搭建难度、原理差异和识别精确度进行对比研究。仿真结果表明,所提方法能够实现对连续多字符中文语音的有效识别,识别正确率在90%以上。     

函数链及其在语音识别中的应用

本文根据加权倒变距离测度，提出了一种用于非特定人语音识别的函数链神经网络。此网络与多层感知器相比，不仅具有较高的识别别率，而且大大缩短了网络的学习时间。我们采用６人（３男，３女）的语音数据对汉语十个数字进行了实验，正确识别率为９３．７％。     

语音识别与理解的研究进展

本文综述了当前语音识别理解的发展趋势和最新进展。指出美国在不依说话人的大词汇表的连续语音隐马尔柯夫模型识别方面起主导地位,日本在大词汇表的连续语音神经网络识别、模拟人工智能进行语音后处理方面起主导地位,并介绍了国际上最优秀的语音识别理解系统。     

Cost-effective test solutions for smart card and other integrated flash applications

From its emergence in the late 1980s as a lower cost alternative to early EEPROM technologies, flash memory has evolved to higher densities and speedsand rapidly growing acceptance in mobile applications.In the process, flash memory devices have placed increased test requirements on manufacturers. Today, as flash device test grows in importance in China, manufacturers face growing pressure for reduced cost-oftest, increased throughput and greater return on investment for test equipment. At the same time, the move to integrated flash packages for contactless smart card applications adds a significant further challenge to manufacturers seeking rapid, low-cost test.     

An improved parallel thinning algorithm with two subiterations

The parallel thinning algorithm with two subiterations is improved in this paper. By analyzing the notions of connected components and passes, a conclusion is drawn that the number of passes and the number of eight-connected components are equal. Then the expression of the number of eight-connected components is obtained which replaces the old one in the algorithm. And a reserving condition is proposed by experiments, which alleviates the excess deletion where a diagonal line and a beeline intersect. The experimental results demonstrate that the thinned curve is almost located in the middle of the original curve connectivelv with single pixel width and the processing speed is high.     

Theoretical analysis of the relationship between the Brillouin gain coefficient and the strain in the optical-fiber sensors

The relation between the power of the Brillouin signal and the strain is one of the bases of the distributed fiber sensors of temperature and strain. The coefficient of the Bfillouin gain can be changed by the temperature and the strain that will affect the power of the Brillouin scattering. The relation between the change of the Brillouin gain coefficient and the strain is thought to be linear by many researchers. However, it is not always linear based on the theoretical analysis and numerical simulation. Therefore, errors will be caused if the relation between the change of the Brillouin gain coefficient and the strain is regarded as to be linear approximately for measuring the temperature and the strain. For this reason, the influence of the parameters on the Brillouin gain coefficient is proposed through theoretical analysis and numerical simulation.     

Novel fiber-optical interferometer with miniaturized probe for in-hole measurements

Today, micro-system technology and the development of new MEMS （Micro-Electro-Mechanical Systems） are emerging rapidly. In order for this development to become a success in the long run, measurement systems have to ensure product quality. Most often, MEMS have to be tested by means of functionality or destructive tests. One reason for this is that there are no suitable systems or sensing probes available which can be used for the measurement of quasi inaccessible features like small holes or cavities. We present a measurement system that could be used for these kinds of measurements. The system combines a fiber optical, miniaturized sensing probe with low-coherence interferometry, so that absolute distance measurements with nanometer accuracy are possible.     

Teleportation of an arbitrary unknown N-qubit entangled state under the controlling of M controllers

A new quantum protocol to teleport an arbitrary unknown N-qubit entangled state from a sender to a fixed receiver under M controllers(M < N) is proposed. The quantum resources required are M non-maximally entangled Greenberger-Home-Zeilinger (GHZ) state and N-M non-maximally entangled Einstein-Podolsky-Rosen (EPR) pairs. The sender performs N generalized Bell-state measurements on the 2N particles. Controllers take M single-particle measurement along x-axis, and the receiver needs to introduce one auxiliary two-level particle to extract quantum information probabilistically with the fidelity unit if controllers cooperate with it.     

Diode-end-pumped Nd:YVO4/LBO lasers with 4.2W continuous-wave output at 457 nm

A continuous-wave (CW) 457 nm blue laser operating at the power of 4.2 W is demonstrated by using a fiber coupled laser diode module pumped Nd: YVO4 and using LBO as the intra-cavity SHG crystal With the optimization of laser cavity and crystal parameters, the laser operates at a very high efficiency. When the pumping power is about 31 W, the output at 457nm reaches 4.2 W, and the optical to optical conversion efficiency is about 13.5% accordingly. The stability of the out putpower is better than 1.2% for 8 h continuously working.     

Call for Papers

正Wireless Body-area Networks The last decade has witnessed the convergence of three giant worlds:electronics,computer science and telecommunications.The next decade should follow this convergence in most of our activities with the generalization of sensor networks.In particular with the progress in medicine,people live longer and the aging of population will push the development of wireless personal networks     

Call for Papers—Feature Topic,Vol.12,No.7,2015

正Information Centric Networking Information-Centric Networking(ICN) is an emerging direction in Future Internet architecture research,gaining significant tractions among academia and industry.Aiming to replace the conventional host-to-host communication model by a data-centric model,ICN treats data content as the first     

JOURNAL OF ELECTRONICS(CHINA)

正~~     

NEXT GENERATION WIRELESS NETWORKS–ESPECIALLY THE USE OF MILLI-MICROWAVE BANDS

The global bandwidth shortage of wireless communications has motivated the exploration of the naillimeter wave （ram-wave） frequency spectrum for the next generation wireless communications. Recent advances in RF CMOS technology and high speed baseband signal processing technologies have enabled tile extensive research and development of turn-wave wireless communications. The multi gigabit per second data rate of ram-wave system will lead to applications in many important scenarios, such as WPAN, WLAN,back-haul for cellular system. And the frequency bands include 28 GHz, 38 GHz, 45GHz, 60GHz, E-BAND and even beyond 100 GHz. The propagation and the imitation of the RF circuits design in these frequency bands make the directional antennas be inevitable for mm-wave communications.