Microphone Array Speech Enhancement System Combining ICA Preprocessing in Highly Noisy Environments
In environments of strong background noise and reverberations, enhancement performance of classic microphone array enhancement systems is still not enough. Since ICA can separate independent sources only from observations, in this paper ICA is applied to highly noisy sensor signals to obtain cleaner speech signals, which are taken as inputs to the following microphone array enhancement systems. Experiments on real data show that the procedure of ICA preprocessing is effective in suppressing ambient noise, and dramatically improving the quality of input signals. Performance of microphone array speech enhancement system combining ICA preprocessing is greatly promoted.
Wanlong Li Ju Liu Jun Du Shuzhong Bai
School of Information Science and Engineering, Shandong University, Jinan 250100, P.R.China School of Information Science and Engineering, Shandong University, Jinan 250100, P.R.China School o
国际会议
2008 International Conference on Audio,Language and Image Processing(2008国际声音、语言、图像过程大会)
镇江
英文
649-652
2008-07-07(万方平台首次上网日期,不代表论文的发表时间)