[1]董添辉[],张玲华[]. 粒子群优化径向基函数网络的语音转换[J].计算机技术与发展,2017,27(05):64-68.
　DONG Tian-hui[],ZHANG Ling-hua[]. Voice Conversion of Radial Basic Function Neural Network of ParticleSwarm Optimization[J].,2017,27(05):64-68.
点击复制

 粒子群优化径向基函数网络的语音转换()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
27
期数:
2017年05期
页码:
64-68
栏目:
智能、算法、系统工程
出版日期:
2017-05-10

文章信息/Info

Title:
 Voice Conversion of Radial Basic Function Neural Network of ParticleSwarm Optimization
文章编号:
1673-629X（2017)05-0064-05
作者:
 董添辉[1]; 张玲华[2]
 1.南京邮电大学 通信与信息工程学院；2.江苏省通信与网络技术工程研究中心
Author(s):
 DONG Tian-hui[1]; ZHANG Ling-hua[2]
关键词:
 ; 语音转换; 径向基函数中心; 改进的粒子群算法; 径向基函数神经网络
Keywords:
 ; voice conversion; centers of RBF; improved particle swarm optimization; radial basis function neural network
分类号:
TN912.3
文献标志码:
A
摘要:
 径向基函数神经网络具有结构简单和学习速度快等特点,因此常被用作语音转换的模型.隐层核函数的中心是影响径向基函数神经网络性能的重要参数,而传统的K-均值聚类算法受初值影响大,全局优化的效果不佳.所以,选择合适的优化算法来调整RBF网络核函数的中心参数,能改善整个网络的性能,从而提升语音转换的效果.而粒子群算法是一种基于迭代的优化算法,具有容易实现、算法参数少、收敛快和突出的全局寻优能力等特点.提出了一种改进的粒子群算法,优化了径向基函数的中心以提高网络性能,便于更准确地获得说话人与目标人之间谱包络的映射关系.实验结果表明,提出的方法能够有效提高神经网络的性能,使转换后的声音更接近于目标声音.
Abstract:
 Due to simple structure and fast learning,Radial Basis Function (RBF) neural network is used commonly in voice conversion system.The center of kernel function in hidden layer is the important parameter of influencing the RBF neural network,but traditional K-means clustering algorithm relies on the initial value,which is ineffective in global optimization.Therefore,it is significance to select a suitable algorithm to modulate the center of function and enhance the effect of voice conversion.Particle swarm algorithm is an optimized one based on iteration,with the characteristics of easy implementation,much less parameters,fast convergence and better global optimization and so on.An improved particle swarm optimization is proposed to optimize the RBF’s centers for improvement of the performance of RBF network,thus enhancing the transformation of speech parameters.The results acquired by modeling and simulation show that the proposed method has effectively improved the performance of neural network and the effect of converted voices is much closer to the goal.

相似文献/References:

[1]翟继友 张鹏.高斯混合模型参数估值算法的优化[J].计算机技术与发展,2011,(11):145.
　ZHAI Ji-you,ZHANG Peng.Optimization of Parameter Estimation Based on Gaussian Mixture Model[J].,2011,(05):145.
[2]李燕萍 张玲华.基于多时间尺度韵律特征分析的语音转换研究[J].计算机技术与发展,2012,(12):67.
　LI Yan-ping,ZHANG Ling-hua.Voice Conversion Research Based on Multi-time Scale Prosodic Feature Analysis[J].,2012,(05):67.
[3]张志宏,吴庆波,邵立松,等.基于飞腾平台TOE协议栈的设计与实现[J].计算机技术与发展,2014,24(07):1.
　ZHANG Zhi-hong,WU Qing-bo,SHAO Li-song,et al. Design and Implementation of TCP/IP Offload Engine Protocol Stack Based on FT Platform[J].,2014,24(05):1.
[4]梁文快,李毅. 改进的基因表达算法对航班优化排序问题研究[J].计算机技术与发展,2014,24(07):5.
　LIANG Wen-kuai,LI Yi. Research on Optimization of Flight Scheduling Problem Based on Improved Gene Expression Algorithm[J].,2014,24(05):5.
[5]黄静,王枫,谢志新,等. EAST文档管理系统的设计与实现[J].计算机技术与发展,2014,24(07):13.
　HUANG Jing,WANG Feng,XIE Zhi-xin,et al. Design and Implementation of EAST Document Management System[J].,2014,24(05):13.
[6]侯善江[],张代远[][][]. 基于样条权函数神经网络P2P流量识别方法[J].计算机技术与发展,2014,24(07):21.
　HOU Shan-jiang[],ZHANG Dai-yuan[][][]. P2P Traffic Identification Based on Spline Weight Function Neural Network[J].,2014,24(05):21.
[7]李璨,耿国华,李康,等. 一种基于三维模型的文物碎片线图生成方法[J].计算机技术与发展,2014,24(07):25.
　LI Can,GENG Guo-hua,LI Kang,et al. A Method of Obtaining Cultural Debris’ s Line Chart Based on Three-dimensional Model[J].,2014,24(05):25.
[8]翁鹤,皮德常. 混沌RBF神经网络异常检测算法[J].计算机技术与发展,2014,24(07):29.
　WENG He,PI De-chang. Chaotic RBF Neural Network Anomaly Detection Algorithm[J].,2014,24(05):29.
[9]刘茜[],荆晓远[],李文倩[],等. 基于流形学习的正交稀疏保留投影[J].计算机技术与发展,2014,24(07):34.
　LIU Qian[],JING Xiao-yuan[,LI Wen-qian[],et al. Orthogonal Sparsity Preserving Projections Based on Manifold Learning[J].,2014,24(05):34.
[10]尚福华,李想,巩淼. 基于模糊框架-产生式知识表示及推理研究[J].计算机技术与发展,2014,24(07):38.
　SHANG Fu-hua,LI Xiang,GONG Miao. Research on Knowledge Representation and Inference Based on Fuzzy Framework-production[J].,2014,24(05):38.

更新日期/Last Update: 2017-07-07