| 基于方差归一化失真测度的语音识别 |
| |
| 引用本文： | 林强,裘雪红.基于方差归一化失真测度的语音识别[J].电子科技,2007(8):39-41. |
| |
| 作者姓名： | 林强 裘雪红 |
| |
| 作者单位： | 西安电子科技大学,计算机学院,陕西,西安,710071 |
| |
| 摘 要： | 在语音识别系统中不同码字有不同的识别能力,识别能力的大小可在权值上加以体现。采用了方差归一化失真测度,将这种有效性差别作为权重矢量反映到失真测度计算公式中,可有效提高基于MFCC(Mel -Frequency Cepstral Coefficients即梅尔倒频谱参数)的语音识别系统的识别性能。
|
| 关 键 词： | 失真测度 方差归一化 矢量量化 MFCC |
| 收稿时间： | 2006-11-13 |
| 修稿时间： | 2006年11月13 |
Voice Recognition Based on Normalized Square-Mean Distortion Measurement |
| |
| Lin Qiang,Qiu Xuehong.Voice Recognition Based on Normalized Square-Mean Distortion Measurement[J].Electronic Science and Technology,2007(8):39-41. |
| |
| Authors: | Lin Qiang Qiu Xuehong |
| |
| Abstract: | Different code words have different validities for voice recognition.Recognition capability can be represented by its weight.This paper introduces square-mean distortion measurement,which can combine the validities of the different distributions with the calculation of the distortion.Experiments show that the dis- tortion measurement can improve the performance of voice recognition system based on MFCC(Mel-Frequency Cepstral Coefficients). |
| |
| Keywords: | distortion measurement normalized square-mean vector quantization MFCC |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
