| 岳喜才,管桦,叶大田.说话人识别使用遗传RBF网络[J].,2000,19(2):35-38 |
| 说话人识别使用遗传RBF网络 |
| Speaker identification using modified radial basis function networks by genetic algorithm |
| |
| 中文摘要: |
| 针对ＲＢＦ网络普遍采用的一种训练算法所存在的局部最佳问题，本文将遗传算法用于ＲＢＦ网络训练过程，增强了该网络的全局寻优能力，提高了ＲＢＦ网络的模式识别性能．说话人识别实验显示，改进训练算法后，ＲＢＦ网络的说话人识别率有稍许提高． |
| 英文摘要: |
| A genetic K-mean clustering algorithm is applied to learning in the hidden layer of radial basis function networks for avoiding only locally optimum solution. The modified radial basis function networks has the ability of global optimization. Experiments showed that the correctness of 10 Chinese speaker identification is improved form 90% to 92%, with 30 seconds speech training and 15 seconds speech test. |
| DOI：10.11684/j.issn.1000-310X.2000.02.009 |
| 中文关键词: 说话人识别 神经网络 遗传算法 |
| 英文关键词: Speaker identification Neural networks Genetic algorithm |
| 基金项目: |
|
| 摘要点击次数: 2434 |
| 全文下载次数: 565 |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
