The role of temporal resolution in modulation-based speech segregation

Tobias May; Thomas Bentsen; Torsten Dau

The role of temporal resolution in modulation-based speech segregation

Tobias May
, Thomas Bentsen
, Torsten Dau

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

374 Downloads (Orbit)

Abstract

This study is concerned with the challenge of automatically segregating a target speech signal from interfering background noise. A computational speech segregation system is presented which exploits logarithmically-scaled amplitude modulation spectrogram (AMS) features to distinguish between speech and noise activity on the basis of individual time-frequency (T-F) units. One important parameter of the segregation system is the window duration of the analysis-synthesis stage, which determines the lower limit of modulation frequencies that can be represented but also the temporal acuity with which the segregation system can manipulate individual T-F units. To clarify the consequences of this trade-off on modulation-based speech segregation performance, the influence of the window duration was systematically investigated

Original language	English
Title of host publication	Proceedings of Interspeech 2015
Number of pages	5
Publication date	2015
Publication status	Published - 2015
Event	INTERSPEECH 2015 : Speech beyond Speech - Dresden, Germany Duration: 6 Sept 2015 → 10 Sept 2015

Conference

Conference	INTERSPEECH 2015
Country/Territory	Germany
City	Dresden
Period	06/09/2015 → 10/09/2015

Keywords

Speech segregation
Ideal binary mask
Amplitude modulation spectrogram features
Temporal resolution

Access to Document

MayBentsenDau Interspeech 2015 SubmittedFinal published version, 402 KB

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{178470a56dc242888220d2038691647b,

title = "The role of temporal resolution in modulation-based speech segregation",

abstract = "This study is concerned with the challenge of automatically segregating a target speech signal from interfering background noise. A computational speech segregation system is presented which exploits logarithmically-scaled amplitude modulation spectrogram (AMS) features to distinguish between speech and noise activity on the basis of individual time-frequency (T-F) units. One important parameter of the segregation system is the window duration of the analysis-synthesis stage, which determines the lower limit of modulation frequencies that can be represented but also the temporal acuity with which the segregation system can manipulate individual T-F units. To clarify the consequences of this trade-off on modulation-based speech segregation performance, the influence of the window duration was systematically investigated",

keywords = "Speech segregation, Ideal binary mask, Amplitude modulation spectrogram features, Temporal resolution",

author = "Tobias May and Thomas Bentsen and Torsten Dau",

year = "2015",

language = "English",

booktitle = "Proceedings of Interspeech 2015",

note = "INTERSPEECH 2015 : Speech beyond Speech ; Conference date: 06-09-2015 Through 10-09-2015",

}

TY - GEN

T1 - The role of temporal resolution in modulation-based speech segregation

AU - May, Tobias

AU - Bentsen, Thomas

AU - Dau, Torsten

PY - 2015

Y1 - 2015

N2 - This study is concerned with the challenge of automatically segregating a target speech signal from interfering background noise. A computational speech segregation system is presented which exploits logarithmically-scaled amplitude modulation spectrogram (AMS) features to distinguish between speech and noise activity on the basis of individual time-frequency (T-F) units. One important parameter of the segregation system is the window duration of the analysis-synthesis stage, which determines the lower limit of modulation frequencies that can be represented but also the temporal acuity with which the segregation system can manipulate individual T-F units. To clarify the consequences of this trade-off on modulation-based speech segregation performance, the influence of the window duration was systematically investigated

AB - This study is concerned with the challenge of automatically segregating a target speech signal from interfering background noise. A computational speech segregation system is presented which exploits logarithmically-scaled amplitude modulation spectrogram (AMS) features to distinguish between speech and noise activity on the basis of individual time-frequency (T-F) units. One important parameter of the segregation system is the window duration of the analysis-synthesis stage, which determines the lower limit of modulation frequencies that can be represented but also the temporal acuity with which the segregation system can manipulate individual T-F units. To clarify the consequences of this trade-off on modulation-based speech segregation performance, the influence of the window duration was systematically investigated

KW - Speech segregation

KW - Ideal binary mask

KW - Amplitude modulation spectrogram features

KW - Temporal resolution

M3 - Article in proceedings

BT - Proceedings of Interspeech 2015

T2 - INTERSPEECH 2015

Y2 - 6 September 2015 through 10 September 2015

ER -

The role of temporal resolution in modulation-based speech segregation

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this