检索结果-南通市图书馆

SOUND SOURCE LOCALIZATION OF DIGITAL HEARING AIDS USING WAVELET BASED MULTIVARIATE STATISTICAL METHOD

在线全文

学校读者我要写书评

暂无评论

Journal of Electronics(China) 2010年第4期27卷 571-576页

作者： liang ruiyu Zou Cairog Wang Qingyu Xi Ji School of Information Science and Engineering Southeast University Nanjing 210096 China College of Computer and Information Hohai University Changzhou 213022 China School of Communication Engineering Nanjing Institute of Technology Nanjing 211167 China

The letter proposed a sound source localization method of digital hearing aids using wavelet based multivariate statistics with the Generalized Cross Correlation (GCC) algorithm. Haar wavelet is used to decompose GCC sequences and extract four wavelet characteristics. And then, Hotelling T2 statistical method is used to fuse the four wavelet characteristics. The statistical value is used to judge the number of sound sources and obtain corresponding time delay estimation which is used to localize the position of sound source. The experimental results show that the proposed method has better robustness in an environment with severe noise and reverberation. Meanwhile, the complexity of al-gorithm is moderate, which is available for sound source localization of hearing aids.

关键词： Sound source localization Wavelet decomposition Hotelling T2 statistical model Digital hearing aids

Research on Nonlinear Frequency Compression Method of Hearing Aid with Adaptive Compression Ratio

在线全文

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2019年第11期121卷 551-567页

作者： Xia Wang Hongming Shen Huawei Tao ruiyu liang Xing Deng Haijian Shao Li Zhao Cairong Zou School of Information Science and Technology Nantong UniversityNantong226019China School of Information Science and Engineering Henan University of TechnologyZhengzhou450001China School of Communication Engineering Nanjing Institute of TechnologyNanjing211167China School of Automation Key Laboratory of Measurement and Control for CSEMinistry of EducationSoutheast UniversityNanjing210096China School of Computer Science Jiangsu University of Science and TechnologyZhenjiang212003China School of Information Science and Engineering Southeast UniversityNanjing210096China

To make full use of the residual audible frequency band of hearing-loss patients and improve the intelligibility of speech,an adaptive nonlinear frequency compression(NFC)algorithm is proposed,which amplifies signals below the cutoff frequency while compresses signals above the cutoff ***,high-frequency signals are decomposed to critical band signals according to the BARK ***,the global compression ratio is determined according to the patient's cutoff frequency and maximum audible ***,the sub-band compression ratio is adaptively determined based on the global compression ratio and normalized average energy of subband ***,the high frequency signals are transposed to low frequency bands by compression mapping,and the phases are adjusted to the same as the original low frequency *** results of speech intelligibility with nine subjects demonstrate that compared to conventional amplitude amplification and nonlinear frequency compression algorithms the proposed algorithm significantly improves the intelligibility of initials and sentences,while not affects the intelligibility of finals and tones significantly.

关键词： Frequency compression sensorineural hearing loss hearing aids

A non-intrusive speech quality evaluation algorithm combining auxiliary target learning and convolutional recurrent network

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Acoustics 2023年第2期42卷 235-250页

作者： TANG Guichen liang ruiyu KONG Fanliu XIE Yue JU Mengjie School of Information and Communication Engineering Nanjing Institute of TechnologyNanjing 211167 School of Information Science and Engineering Southeast UniversityNanjing 210096

The objective evaluation of speech quality can replace expensive manual scoring,but current objective indicators usually need pure reference speech,which is difficult to obtain in many practical acoustic systems.A non-intrusive speech quality evaluation algorithm combining auxiliary target learning and a convolutional recurrent network(CRN)is *** frequency cepstral coefficients(BFCCs),which are based on human-like auditory filters,are used as the input of the CRN network to effectively reduce the network ***,frame-level features are extracted by a convolutional neural network(CNN)from ***,long-term time dependence and sequence features are modeled by the bidirectional long shortterm memory(BiLSTM)networks in frame-level ***,a self-attention mechanism is introduced into the CRN,thereby adaptively extracting useful information from frame-level features,which is then integrated into the characteristics of the sentence level and mapped into the final objective *** addition,a multi-task training strategy is adopted,and voice activity detection(VAD)is introduced as an auxiliary learning target to improve the performance of the *** in public databases show that compared with other non-intrusive algorithms,the proposed algorithm has a better correlation with the mean opinion score(MOS).Moreover,it has a small parameter size and good generalization ability for the distorted speech database with MOS released by ITU-T P.808,which is close to the accuracy of the perceptual evaluation of speech quality(PESQ).

关键词： network speech algorithm

Transfer learning with deep sparse auto-encoder for speech emotion recognition

在线全文

学校读者我要写书评

暂无评论

Journal of Southeast University(English Edition) 2019年第2期35卷 160-167页

作者： liang Zhenlin liang ruiyu Tang Manting Xie Yue Zhao Li Wang Shijia School of Information Science and Engineering Southeast UniversityNanjing 210096China School of Communication Engineering Nanjing Institute of TechnologyNanjing 211167China School of Computer Engineering Jinling Institute of TechnologyNanjing 211169China

In order to improve the efficiency of speech emotion recognition across corpora,a speech emotion transfer learning method based on the deep sparse auto-encoder is *** algorithm first reconstructs a small amount of data in the target domain by training the deep sparse auto-encoder,so that the encoder can learn the low-dimensional structural representation of the target domain ***,the source domain data and the target domain data are coded by the trained deep sparse auto-encoder to obtain the reconstruction data of the low-dimensional structural representation close to the target ***,a part of the reconstructed tagged target domain data is mixed with the reconstructed source domain data to jointly train the *** part of the target domain data is used to guide the source domain *** on the CASIA,SoutheastLab corpus show that the model recognition rate after a small amount of data transferred reached 89.2%and 72.4%on the *** to the training results of the complete original corpus,it only decreased by 2%in the CASIA corpus,and only 3.4%in the SoutheastLab *** show that the algorithm can achieve the effect of labeling all data in the extreme case that the data set has only a small amount of data tagged.

关键词： sparse auto-encoder transfer learning speech emotion recognition

基于BP神经网络的采空区激光探测环境误差修正

在线全文

学校读者我要写书评

暂无评论

中南大学学报（自然科学版） 2020年第3期51卷 758-766页

作者：侯朋远徐帅梁瑞余纪晓飞东北大学深部金属矿山安全开采教育部重点实验室辽宁沈阳110819 赤峰山金红岭有色矿业有限责任公司内蒙古赤峰025450

为了研究地下金属矿山高温?高湿?多尘复杂环境对采空区三维激光探测精度的影响,研发采空区复杂环境模拟装置,设计64组正交试验方案,分析温度、相对湿度和粉尘质量浓度对探测误差的影响并进行修正。为进一步量化探测误差,建立BP神经网络... 详细信息

为了研究地下金属矿山高温?高湿?多尘复杂环境对采空区三维激光探测精度的影响,研发采空区复杂环境模拟装置,设计64组正交试验方案,分析温度、相对湿度和粉尘质量浓度对探测误差的影响并进行修正。为进一步量化探测误差,建立BP神经网络预测模型,并应用于某铅锌矿采空区的探测中。研究结果表明:随相对湿度升高,点云平均误差比呈“S”型曲线增长,当相对湿度介于75.0%~85.0%时,增长速率明显加快;随粉尘质量浓度增大,点云平均误差比以40 mg/m3和100 mg/m3为节点呈三段式线性增长,当粉尘质量浓度介于40~100 mg/m3时,增长速率最快,当粉尘质量浓度大于100 mg/m3时,增长速率次之,当粉尘质量浓度小于40 mg/m3时,增长速率最低;温度对点云平均误差比影响较小。模型预测值与实测值的平均相对误差为1.80%,校正决定系数为0.993;修正后的采空区边界标高、体积和顶板暴露面积等参数更加符合实际情况。

关键词：复杂环境采空区三维激光探测正交试验 BP神经网络误差分析误差修正

在线全文

学校读者我要写书评

暂无评论

数字助听器语音处理算法研究进展与展望

数据采集与处理 2016年第2期31卷 242-251页

作者：邹采荣梁瑞宇谢跃广州大学机械与电器工程学院广州510006 东南大学信息科学工程学院南京210096 南京工程学院通信工程学院南京211167

全球老龄化使听力障碍成为高发性慢性疾病,而佩戴助听器是老年性聋患者听力干预和康复最有效的手段之一。随着数字信号处理技术和电子技术的飞速进步,近年来应用于数字助听器的各种算法和技术得到了显著的发展。其中声场景分类、滤波器... 详细信息

全球老龄化使听力障碍成为高发性慢性疾病,而佩戴助听器是老年性聋患者听力干预和康复最有效的手段之一。随着数字信号处理技术和电子技术的飞速进步,近年来应用于数字助听器的各种算法和技术得到了显著的发展。其中声场景分类、滤波器分解、噪声抑制和回声消除是助听器的4个基本算法。基于对这些算法的研究,本文从算法基本原理、当前研究现状、算法特点以及存在的问题进行分析介绍。此外,通过分析现阶段数字助听器算法中存在的问题,介绍了3个助听器信号处理方面的最新研究方向——听觉仿生、听觉认知和自验配助听器。本文最后对未来研究进行了展望。

关键词：数字助听器场景分类滤波器分解回声消除噪声抑制

在线全文

学校读者我要写书评

暂无评论

基于改进LSTM的儿童语音情感识别模型

计算机工程 2020年第6期46卷 40-49页

作者：余莉萍梁镇麟梁瑞宇复旦大学计算机科学技术学院上海201203 东南大学信息科学工程学院南京210096 南京工程学院信息与通信工程学院南京211167

为实现不同儿童情感需求状态下帧级语音特征的有效获取,建立一种基于改进长短时记忆(LSTM)网络的儿童语音情感识别模型。采用帧级语音特征代替传统统计特征以保留原始语音中的时序关系,通过引入注意力机制将传统遗忘门和输入门转换为注... 详细信息

为实现不同儿童情感需求状态下帧级语音特征的有效获取,建立一种基于改进长短时记忆(LSTM)网络的儿童语音情感识别模型。采用帧级语音特征代替传统统计特征以保留原始语音中的时序关系,通过引入注意力机制将传统遗忘门和输入门转换为注意力门,并根据自定义的深度策略计算得到深度注意力门,从而提高语音情感识别性能。实验结果表明,在Fau Aibo儿童情感数据语料库及婴儿哭声情感需求数据库上,该模型在召回率和F1分数上相比基于传统LSTM的识别模型分别提高了3.14%、5.50%和1.84%、5.49%,在CASIA中文情感数据库上,其相比基于传统LSTM和GRU的识别模型训练时间更短、儿童语音情感识别率更高。

关键词：儿童情感时序关系帧级语音特征深度注意力门长短时记忆网络

改进的FastICA算法在图像盲分离中的应用

在线全文

学校读者我要写书评

暂无评论

电子技术应用 2013年第1期39卷 116-118页

作者：阴法明梁瑞宇南京信息职业技术学院江苏南京210046 河海大学江苏常州213022

在系统研究独立分量分析基本原理、快速算法的基础上,提出了一种改进的独立分量分析快速算法。引入了一种综合考虑峰度和偏度的新的简单目标函数,并对独立分量分析的快速算法流程提出了改进。改进的算法在图像盲分离中得到了应用。实验... 详细信息

在系统研究独立分量分析基本原理、快速算法的基础上,提出了一种改进的独立分量分析快速算法。引入了一种综合考虑峰度和偏度的新的简单目标函数,并对独立分量分析的快速算法流程提出了改进。改进的算法在图像盲分离中得到了应用。实验结果表明,该算法显著提高了独立分量分离效果。

关键词：独立分量分析峰度偏度图像盲分离

融合改进梅尔谱特征和深信念网络的语音测谎算法

在线全文

学校读者我要写书评

暂无评论

声学学报 2019年第2期44卷 214-220页

作者：谢跃梁瑞宇包永强赵力王诗佳东南大学信息科学工程学院南京210096 南京工程学院通信工程学院南京211167

为实现非接触式谎言检测,特提出了以语谱特征为线索,结合深度学习的谎言检测方法。为提取谎言中微颤抖所引起的语谱局部能量变化,算法先对梅尔频谱进行了Hu矩处理,然后进行离散余弦变换去除相关性。该特征利用了Hu矩的正交不变性和平移... 详细信息

为实现非接触式谎言检测,特提出了以语谱特征为线索,结合深度学习的谎言检测方法。为提取谎言中微颤抖所引起的语谱局部能量变化,算法先对梅尔频谱进行了Hu矩处理,然后进行离散余弦变换去除相关性。该特征利用了Hu矩的正交不变性和平移不变性,能较好的体现出语谱中局部能量的集中方式。然后将所提取的特征作为改进深信念网络输入进行谎言识别。为提高受限玻尔兹曼机的并行回火训练算法中相邻温度链之间的交换率,训练算法先对Markov链的状态能量进行等能量的划分,使得每个能量环内的状态具有相似的能量,然后再进行交换以提高交换率从而优化整个网络的训练。在Columbia-SRI-Colorado数据库上的实验表明,谎言识别率达到了71.47%,比梅尔倒谱系数特征的识别率提高了3%,比传统的BayesNet分类算法提高了7%。

关键词：算法比较训练算法识别率语音段基音频率 DBN 谱特征 RBM