| 基于粒子群优化神经网络的语音情感识别 |
| |
| 引用本文： | 余华,黄程韦,金赟,赵力.基于粒子群优化神经网络的语音情感识别[J].数据采集与处理,2011,26(1). |
| |
| 作者姓名： | 余华 黄程韦 金赟 赵力 |
| |
| 作者单位： | 1. 南京信息职业技术学院电子信息学院,南京,210013;东南大学信息科学与工程学院,南京,210096
2. 东南大学信息科学与工程学院,南京,210096 |
| |
| 基金项目： | 国家自然科学基金(60472058,60975017)资助项目; 江苏省自然科学基金(BK200829)资助项目 |
| |
| 摘 要： | 提出了一种基于粒子群优化算法的人工神经网络,并把它应用到语音情感识别系统中。依据情感的维度空间模型,分别提取了韵律特征与音质特征,研究了谐波噪声比特征随情感类别的变化。利用粒子群优化算法(PSO)训练随机产生的初始数据,优化神经网络的连接权值和阈值,快速地实现网络的收敛。在实验中比较了BP神经网络、RBF神经网络与PSO神经网络分别用于语音情感识别的识别率,PSO神经网络的平均识别率高于BP神经网络6.7%,高于RBF神经网络5.4%。结果显示,粒子群优化神经网络用于语音情感识别提高了识别性能。
|
| 关 键 词： | 语音情感识别 粒子群优化 神经网络 |
Speech Emotion Recognition Based on Particle Swarm Optimizer Neural Network |
| |
| Yu Hua,Huang Chengwei,Jin Yun,Zhao Li.Speech Emotion Recognition Based on Particle Swarm Optimizer Neural Network[J].Journal of Data Acquisition & Processing,2011,26(1). |
| |
| Authors: | Yu Hua Huang Chengwei Jin Yun Zhao Li |
| |
| Affiliation: | Yu Hua1,2,Huang Chengwei2,Jin Yun2,Zhao Li2(1.Electronic Information College,Nanjing Information Occupation Technical College,Nanjing,210013,China,2.School of Information Science and Engineering,Southeast University,210096,China) |
| |
| Abstract: | Particle swarm optimizer(PSO) neural network is used in speech emotion recognition system.According to the dimensional model,emotion features are extracted and categorized into prosody features and voice quality features.HNR feature is studied corresponding to different emotion categories.The particle swarm optimizer is used to train the random initial data to optimize the connection weights and thresholds of the neural network,thus having a fast network convergence speed.In the recognition experiments,BP n... |
| |
| Keywords: | speech emotion recognition particle swarm optimizer neural network |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
