Publication detail

Blind source separation using fuzzy masking

ČERMÁK, J.

Original Title

Blind source separation using fuzzy masking

Type

conference paper

Language

English

Original Abstract

This paper deals with separation of the source signals from their mixtures without any information about mixing environment. This problem is known as a blind source separation. In this paper we focus on improving a multichannel method called time-frequency binary masking. Our proposed approach is based on using a fuzzy mask instead of a binary mask. We will show that the separation performance increases by employing fuzzy mask. Moreover our method reduces the musical noise introduced by binary mask to the separated signals.

Keywords

speech processing, signal processing, source separation

Authors

ČERMÁK, J.

RIV year

2007

Released

1. 12. 2007

Publisher

UFE AVCR

Location

Praque

ISBN

978-80-86269-00-9

Book

SPEECH PROCESSING 17th Czech – German Workshop

Pages from

83

Pages to

87

Pages count

5

BibTex

@inproceedings{BUT25567,
  author="Jan {Čermák}",
  title="Blind source separation using fuzzy masking",
  booktitle="SPEECH PROCESSING 17th Czech – German Workshop",
  year="2007",
  pages="83--87",
  publisher="UFE AVCR",
  address="Praque",
  isbn="978-80-86269-00-9"
}