| Paper: | SLP-P11.4 | ||
| Session: | Front-end For Robust Speech Recognition | ||
| Time: | Wednesday, May 17, 16:30 - 18:30 | ||
| Presentation: | Poster | ||
| Topic: | Speech and Spoken Language Processing: Feature-based Robust Speech Recognition (e.g., noise, etc) | ||
| Title: | A NOVEL NOISE ROBUST FRONT-END USING FIRST ORDER VTS IN CONSTRUCTION OF MEL-WARPED WIENER FILTER | ||
| Authors: | Mu Su, Peng Li, Zhuo Wang, Peng Ding, Bo Xu, Institute of Automation, China | ||
| Abstract: | In this paper, we first review two approaches in the context of robust recognition, e.g. speech enhancement based two-stage mel warp wiener filtering (MWF) and first-order Vector Taylor Series (VTS)compensation in log power spectrum, which are widely used. A new algorithm is pro-posed which combined these two approaches in a way, where VTS compensation derived statistics are used to con-struct the wiener filter. We will show that this noise robust front end is superior. The experiments results prove that our proposed method does show significant advantage over VTS and MWF. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012