2008•International Conference on Signal ProcessingRequires access

Performance analysis of single channel blind source separation

Bin Gao, Wai Lok Woo, S.S. Dlay

Abstract

Blind source separation is an advanced statistical tool that has found widespread use in many signal processing applications. The main idea to single channel blind source separation is based on exploiting the inherent time structure of sources known as basis filters in time domain that encode the sources in a statistically efficient manner. This paper is focused on the analysis of separation results by exploiting one channel recording of real time speech and music signals.

About this research paper

What this paper is about

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Key concepts: Blind signal separation, Source separation, Computer science, Independent component analysis, Channel (broadcasting), Signal processing, Speech recognition, Separation (statistics)