Combining standard and throat microphones for robust speech recognition

被引：67

作者：

Graciarena, M

Franco, H

Sonmez, K

Bratt, H

机构：

[1] SRI Int, Speech Technol & Res Lab, Menlo Pk, CA 94025 USA

[2] Univ Buenos Aires, Sch Engn, Inst Biomed Engn, RA-1053 Buenos Aires, DF, Argentina

来源：

IEEE SIGNAL PROCESSING LETTERS | 2003年 / 10卷 / 03期

关键词：

noise robustness; probabilistic optimum filtering; speech recognition; throat microphone;

D O I：

10.1109/LSP.2003.808549

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

We present a method to combine the standard and throat microphone signals for robust speech recognition in noisy environments. Our approach is to use the. probabilistic optimum filter (POF) mapping algorithm to estimate the standard microphone clean-speech feature vectors, used by standard speech recognizers, from both microphones' noisy-speech feature vectors. A small untranscribed "stereo" database (noisy and clean simultaneous recordings) is required to train the POF mappings. In continuous-speech recognition experiments using SRI International's DECIPHER recognition system, both using artificially added noise and using recorded noisy speech, the combined-microphone approach significantly outperforms the single-microphone approach.

引用

页码：72 / 74

页数：3