Feature extraction and model-based noise compensation for noisy speech recognition evaluated on AURORA 2 task

Kaisheng Yao, Jingdong Chen, Kuldip K. Paliwal, Satoshi Nakamura

科研成果: 书/报告/会议事项章节会议稿件同行评审

4 引用 (Scopus)

摘要

We have evaluated several feature-based and a model-based method for robust speech recognition in noise. The evaluation was performed on Aurora 2 task. We show that after a subband based spectral subtraction, features can be more robust to additive noise. We also report a robust feature set derived from differential power spectrum (DPS), which is not only robust to additive noise, but also robust to spectrum colorization due to channel effects. When the clean training set is available, we show that a model-based noise compensation method can be effective to improve system robustness to noise. Given the testing sets, as a whole, the feature-based methods can yield about 22% relative improvement in accuracy for multi-condition training task, and the model-based method can have about 63% relative performance improvement when systems were trained on clean training set.

源语言英语
主期刊名EUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology
编辑Borge Lindberg, Henrik Benner, Paul Dalsgaard, Zheng-Hua Tan
出版商International Speech Communication Association
233-236
页数4
ISBN(电子版)8790834100, 9788790834104
出版状态已出版 - 2001
已对外发布
活动7th European Conference on Speech Communication and Technology - Scandinavia, EUROSPEECH 2001 - Aalborg, 丹麦
期限: 3 9月 20017 9月 2001

出版系列

姓名EUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology

会议

会议7th European Conference on Speech Communication and Technology - Scandinavia, EUROSPEECH 2001
国家/地区丹麦
Aalborg
时期3/09/017/09/01

指纹

探究 'Feature extraction and model-based noise compensation for noisy speech recognition evaluated on AURORA 2 task' 的科研主题。它们共同构成独一无二的指纹。

引用此