Skip to main navigation Skip to search Skip to main content

Blind separation of speech sources in multichannel compressed sensing

  • Harbin Institute of Technology
  • Beijing Institute of Remote Sensing Equipment

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper presents a novel framework for separating and reconstructing multichannel speech sources from compressively sensed linear mixtures simultaneously. The conventional approaches for blind speech separation are almost based on the Nyquist sampling theory. We proposed an approach which uses the multichannel compressive sensing theory for blind speech separation. The linear programming and gradient-based methods are used to separate the sources. Compared with the conventional blind speech separation, the proposed approach can reduce the requirements of sampling speed and operating rate of the devices. Moreover, our approach has lower computational complexity. The main contribution of this paper lies in proposing a novel procedure to estimate the sources from the measurements without reconstructing the mixed signals. Simulation results demonstrate the proposed algorithm can separate multichannel speech sources successfully.

Original languageEnglish
Title of host publication2013 IEEE International Instrumentation and Measurement Technology Conference
Subtitle of host publicationInstrumentation and Measurement for Life, I2MTC 2013 - Proceedings
Pages1771-1774
Number of pages4
DOIs
StatePublished - 2013
Event2013 IEEE International Instrumentation and Measurement Technology Conference: Instrumentation and Measurement for Life, I2MTC 2013 - Minneapolis, MN, United States
Duration: 6 May 20139 May 2013

Publication series

NameConference Record - IEEE Instrumentation and Measurement Technology Conference
ISSN (Print)1091-5281

Conference

Conference2013 IEEE International Instrumentation and Measurement Technology Conference: Instrumentation and Measurement for Life, I2MTC 2013
Country/TerritoryUnited States
CityMinneapolis, MN
Period6/05/139/05/13

Keywords

  • blind source separation
  • compressed sensing
  • mixture model
  • sparse approximation

Fingerprint

Dive into the research topics of 'Blind separation of speech sources in multichannel compressed sensing'. Together they form a unique fingerprint.

Cite this