| 实时识别中语音的时域─频域─能量表示 |
| |
| 引用本文： | 罗万伯,张露,李志蜀,罗霄峰,罗霄岚,Duanpei Wu.实时识别中语音的时域─频域─能量表示[J].应用声学,2001,20(3):31-33,10. |
| |
| 作者姓名： | 罗万伯 张露 李志蜀 罗霄峰 罗霄岚 Duanpei Wu |
| |
| 作者单位： | 1. 四川大学
2. SONY 美国公司 |
| |
| 摘 要： | 本文提出了语音信号的一种时域－频域－能量表示，并给出了算法，可用于孤立词语音识别，这种时域－频域－能量表示有两个特点，基于短时能量梯度的非线性时间规正，可保留语音信号频域的过滤特性，丢掉其稳态特性，计算量小，适于实时应用。
|
| 关 键 词： | 语音识别 语音信号 非线性时间规正 实时实现 时域 频域 能量 |
The time frequency-energy representation off speech signal in real-time recognition |
| |
| Luo Wanbo,Zhang lu,Li Zhishu,Luo Xiaofeng,Luo Xiaolan and Duanpei Wu.The time frequency-energy representation off speech signal in real-time recognition[J].Applied Acoustics,2001,20(3):31-33,10. |
| |
| Authors: | Luo Wanbo Zhang lu Li Zhishu Luo Xiaofeng Luo Xiaolan and Duanpei Wu |
| |
| Institution: | Sichuan University, Chengdu 610064;Sichuan University, Chengdu 610064;Sichuan University, Chengdu 610064;Sichuan University, Chengdu 610064;Sichuan University, Chengdu 610064;SONY Inc., USA |
| |
| Abstract: | A time-frequency-energy representation of speech signal and its algorithm are introduced to isolated-word speech recognition. It can be characterized by two aspects: (1) non-linear time normalization, based on the gradients of short-time energy in a specific number of frequency bands, retains the transient portions and ignores the steady-state portions of speech signal in frequency domain. (2) real-time implementation due to low computational load. |
| |
| Keywords: | Speech recognition Speech signal Dynamic time warping Real-time implementation |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《应用声学》浏览原始摘要信息 |
| 点击此处可从《应用声学》下载免费的PDF全文 |
|
