Binaural detection of speech sources in complex acoustic scenes

Tobias May; Steven Van De Par; Armin Kohlrausch

doi:10.1109/ASPAA.2011.6082276

Binaural detection of speech sources in complex acoustic scenes

Tobias May
, Steven Van De Par
, Armin Kohlrausch

Eindhoven University of Technology
University of Oldenburg

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

204 Downloads (Orbit)

Abstract

In this paper we present a novel system that is able to simultaneously localize and detect a predefined number of speech sources in complex acoustic scenes based on binaural signals. The system operates in two steps: First, the acoustic scene is analyzed by a binaural front-end that detects relevant sound source activity. Second, a speech detection module selects source positions from a set of candidate positions that are most likely speech. The proposed method is evaluated in simulated multi-source scenarios consisting of two speech sources, three interfering noise sources and reverberation. © 2011 IEEE.

Original language	English
Title of host publication	2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics
Publisher	IEEE
Publication date	2011
DOIs	https://doi.org/10.1109/ASPAA.2011.6082276
Publication status	Published - 2011
Externally published	Yes
Event	2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics - New Paltz, United States Duration: 16 Oct 2011 → 19 Oct 2011

Conference

Conference	2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics
Country/Territory	United States
City	New Paltz
Period	16/10/2011 → 19/10/2011

Access to Document

10.1109/ASPAA.2011.6082276

06082276Final published version, 948 KB

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{b56167ba50fa47a894b5ffce197c0b1d,

title = "Binaural detection of speech sources in complex acoustic scenes",

abstract = "In this paper we present a novel system that is able to simultaneously localize and detect a predefined number of speech sources in complex acoustic scenes based on binaural signals. The system operates in two steps: First, the acoustic scene is analyzed by a binaural front-end that detects relevant sound source activity. Second, a speech detection module selects source positions from a set of candidate positions that are most likely speech. The proposed method is evaluated in simulated multi-source scenarios consisting of two speech sources, three interfering noise sources and reverberation. {\textcopyright} 2011 IEEE.",

author = "Tobias May and Par, \{Steven Van De\} and Armin Kohlrausch",

year = "2011",

doi = "10.1109/ASPAA.2011.6082276",

language = "English",

booktitle = "2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics",

publisher = "IEEE",

address = "United States",

note = "2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA 2011 ; Conference date: 16-10-2011 Through 19-10-2011",

}

TY - GEN

T1 - Binaural detection of speech sources in complex acoustic scenes

AU - May, Tobias

AU - Par, Steven Van De

AU - Kohlrausch, Armin

PY - 2011

Y1 - 2011

N2 - In this paper we present a novel system that is able to simultaneously localize and detect a predefined number of speech sources in complex acoustic scenes based on binaural signals. The system operates in two steps: First, the acoustic scene is analyzed by a binaural front-end that detects relevant sound source activity. Second, a speech detection module selects source positions from a set of candidate positions that are most likely speech. The proposed method is evaluated in simulated multi-source scenarios consisting of two speech sources, three interfering noise sources and reverberation. © 2011 IEEE.

AB - In this paper we present a novel system that is able to simultaneously localize and detect a predefined number of speech sources in complex acoustic scenes based on binaural signals. The system operates in two steps: First, the acoustic scene is analyzed by a binaural front-end that detects relevant sound source activity. Second, a speech detection module selects source positions from a set of candidate positions that are most likely speech. The proposed method is evaluated in simulated multi-source scenarios consisting of two speech sources, three interfering noise sources and reverberation. © 2011 IEEE.

U2 - 10.1109/ASPAA.2011.6082276

DO - 10.1109/ASPAA.2011.6082276

M3 - Article in proceedings

BT - 2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics

PB - IEEE

T2 - 2011 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics

Y2 - 16 October 2011 through 19 October 2011

ER -

Binaural detection of speech sources in complex acoustic scenes

Abstract

Conference

Access to Document

OpenUrl availability

Fingerprint

Cite this