摘要
阐述了在噪声条件下,将基于压缩感知理论的丢失数据重建技术应用于说话人识别系统的系统前端。首先使用Mel滤波器组将带噪语音信号转换成Mel频谱,然后利用带噪Mel谱中可靠数据重建不可靠数据,最后从重建的Mel频谱中提取Mel倒谱特征参数用于说话人识别。稳健性实验结果表明,该方法能够提高在噪声环境下说话人系统的识别率。
In this paper, the method of missing data imputation based on the emergent field of compressive sensing for the front end of a speaker recognition system in noisy conditions is investigated. Firstly, noisy speech signals are transformed into Mel spectrum by using Mel filtering. Then, unreliable spectral components are reconstructed given an incomplete set of reliable ones. Finally, speaker features with auditory model are extracted from reconstructed Mel-spectral data. Experimental results demonstrate that the method can improve the identification accuracy of speaker recognition in noisy environments.
出处
《电声技术》
2011年第2期61-63,共3页
Audio Engineering
关键词
压缩感知
缺失数据重建
Mel频谱
说话人识别
compressive sensing
missing date imputation
Mel spectrum
speaker recognition
