JAIST Repository >
School of Information Science >
Articles >
Journal Articles >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10119/9951

Title: Towards intelligent binaural speech enhancement by meaningful sound extraction
Authors: Chau, Duc Thanh
Li, Junfeng
Akagi, Masato
Issue Date: 2011-07
Publisher: Research Institute of Signal Processing Japan
Magazine name: Journal of signal processing
Volume: 15
Number: 4
Start page: 291
End page: 294
Abstract: Current speech enhancement applications, such as binaural hearing aids, mainly aim at suppressing interference signals and preserving the target signal with its binaural cues. However, in addition to the target signal, human beings are able to catchother important or meaningful sounds (e.g., calls from others) in daily conversation. This attention mechanism. to meaningful signals is seldom taken into consideration in state-of-the-art signalprocessing systems. In this paper, we propose an intelligent binaural speech enhancement model by extracting the meaningful signals as well as enhancing the target signal. In particular, the proposed model consists of two main parallel processes: binaural target signal enhancement and binaural meaningful signal extraction, and finally yields the binaural outputs. Experimental results showed that the proposed system is able to notonly suppress the interfering noise signals, but also enhance the target signal and non-target meaningful signals.
Rights: Copyright (C) 2011 Research Institute of Signal Processing Japan. Duc Thanh Chau, Junfeng Li and Masato Akagi, Journal of signal processing, 15(4), 2011, 291-294.
URI: http://hdl.handle.net/10119/9951
Material Type: publisher
Appears in Collections:b10-1. 雑誌掲載論文 (Journal Articles)

Files in This Item:

File Description SizeFormat
JSP2011_Duc.pdf1563KbAdobe PDFView/Open

All items in DSpace are protected by copyright, with all rights reserved.

 


Contact : Library Information Section, Japan Advanced Institute of Science and Technology