Noise estimation for speech enhancement using minimum-spectral-average and vowel-presence detection approach

Ching Ta Lu, Yung Yue Chen, Jun Hong Shen, Ling Ling Wang, Chung Lin Lei

研究成果: Conference contribution

摘要

The accuracy of noise estimation is important for the performance of a speech enhancement system. This study proposes using variable segment length for noise tracking and variable thresholds for the determination of speech-presence probability. Initially, the fundamental frequency is estimated to determine whether a frame is a vowel. In the case of a vowel frame, the segment length increases; meanwhile the threshold for speech-presence is decreased. So the noise magnitude is adequately underestimated. The speech distortion is accordingly reduced in enhanced speech. Conversely, the segment length is rapidly decreased during noise-dominant regions. This enables the noise estimate to be updated quickly and the noise variation to be well tracked, yielding background noise being efficiently removed by the process of speech enhancement. Experimental results show that the proposed method can efficiently track the variation of background noise, enabling the performance of speech enhancement to be improved.

原文English
主出版物標題Frontier Computing - Theory, Technologies and Applications, FC 2016
編輯Neil Y. Yen, Jason C. Hung
發行者Springer Verlag
頁面317-327
頁數11
ISBN(列印)9789811031861
DOIs
出版狀態Published - 2018
事件 5th International Conference on Frontier Computing, FC 2016 - Tokyo, Japan
持續時間: 2016 七月 132016 七月 15

出版系列

名字Lecture Notes in Electrical Engineering
422
ISSN(列印)1876-1100
ISSN(電子)1876-1119

Other

Other 5th International Conference on Frontier Computing, FC 2016
國家Japan
城市Tokyo
期間16-07-1316-07-15

All Science Journal Classification (ASJC) codes

  • Industrial and Manufacturing Engineering

指紋 深入研究「Noise estimation for speech enhancement using minimum-spectral-average and vowel-presence detection approach」主題。共同形成了獨特的指紋。

引用此