Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization

Jürgen T. Geiger; Ravichander Vipperla; Simon Bozonnet; Nicholas Evans; Björn Schuller; Gerhard Rigoll

Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization

Jürgen T. Geiger, Ravichander Vipperla, Simon Bozonnet, Nicholas Evans, Björn Schuller, Gerhard Rigoll

Publikation: Beitrag in Buch/Bericht/Konferenzband › Konferenzbeitrag › Begutachtung

14 Zitate (Scopus)

Abstract

The effective handling of overlapping speech is at the limits of the current state of the art in speaker diarization. This paper presents our latest work in overlap detection. We report the combination of features derived through convolutive nonnegative sparse coding and new energy, spectral and voicingrelated features within a conventional HMM system. Overlap detection results are fully integrated into our top-down diarization system through the application of overlap exclusion and overlap labeling. Experiments on a subset of the AMI corpus show that the new system delivers significant reductions in missed speech and speaker error. Through overlap exclusion and labelling the overall diarization error rate is shown to improve by 6.4 % relative.

Originalsprache	Englisch
Titel	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Seiten	2151-2154
Seitenumfang	4
Publikationsstatus	Veröffentlicht - 2012
Veranstaltung	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 - Portland, OR, USA/Vereinigte Staaten Dauer: 9 Sept. 2012 → 13 Sept. 2012

Publikationsreihe

Name	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Band	3

Konferenz

Konferenz	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Land/Gebiet	USA/Vereinigte Staaten
Ort	Portland, OR
Zeitraum	9/09/12 → 13/09/12

Andere Dateien und Links

Verknüpfung zur Publikation in Scopus

Dieses zitieren

Geiger, J. T., Vipperla, R., Bozonnet, S., Evans, N., Schuller, B., & Rigoll, G. (2012). Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization. in 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 (S. 2151-2154). (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012; Band 3).

Geiger, Jürgen T. ; Vipperla, Ravichander ; Bozonnet, Simon et al. / Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 2012. S. 2151-2154 (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012).

@inproceedings{15c06dbbd81f4455938690d3a840870a,

title = "Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization",

abstract = "The effective handling of overlapping speech is at the limits of the current state of the art in speaker diarization. This paper presents our latest work in overlap detection. We report the combination of features derived through convolutive nonnegative sparse coding and new energy, spectral and voicingrelated features within a conventional HMM system. Overlap detection results are fully integrated into our top-down diarization system through the application of overlap exclusion and overlap labeling. Experiments on a subset of the AMI corpus show that the new system delivers significant reductions in missed speech and speaker error. Through overlap exclusion and labelling the overall diarization error rate is shown to improve by 6.4 % relative.",

keywords = "Convolutive nonnegative sparse coding, Speaker diarization, Speech overlap detection",

author = "Geiger, {J{\"u}rgen T.} and Ravichander Vipperla and Simon Bozonnet and Nicholas Evans and Bj{\"o}rn Schuller and Gerhard Rigoll",

year = "2012",

language = "English",

isbn = "9781622767595",

series = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012",

pages = "2151--2154",

booktitle = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012",

note = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 ; Conference date: 09-09-2012 Through 13-09-2012",

}

Geiger, JT, Vipperla, R, Bozonnet, S, Evans, N, Schuller, B & Rigoll, G 2012, Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization. in 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Bd. 3, S. 2151-2154, 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Portland, OR, USA/Vereinigte Staaten, 9/09/12.

Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization. / Geiger, Jürgen T.; Vipperla, Ravichander; Bozonnet, Simon et al.
13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 2012. S. 2151-2154 (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012; Band 3).

Publikation: Beitrag in Buch/Bericht/Konferenzband › Konferenzbeitrag › Begutachtung

TY - GEN

T1 - Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization

AU - Geiger, Jürgen T.

AU - Vipperla, Ravichander

AU - Bozonnet, Simon

AU - Evans, Nicholas

AU - Schuller, Björn

AU - Rigoll, Gerhard

PY - 2012

Y1 - 2012

N2 - The effective handling of overlapping speech is at the limits of the current state of the art in speaker diarization. This paper presents our latest work in overlap detection. We report the combination of features derived through convolutive nonnegative sparse coding and new energy, spectral and voicingrelated features within a conventional HMM system. Overlap detection results are fully integrated into our top-down diarization system through the application of overlap exclusion and overlap labeling. Experiments on a subset of the AMI corpus show that the new system delivers significant reductions in missed speech and speaker error. Through overlap exclusion and labelling the overall diarization error rate is shown to improve by 6.4 % relative.

AB - The effective handling of overlapping speech is at the limits of the current state of the art in speaker diarization. This paper presents our latest work in overlap detection. We report the combination of features derived through convolutive nonnegative sparse coding and new energy, spectral and voicingrelated features within a conventional HMM system. Overlap detection results are fully integrated into our top-down diarization system through the application of overlap exclusion and overlap labeling. Experiments on a subset of the AMI corpus show that the new system delivers significant reductions in missed speech and speaker error. Through overlap exclusion and labelling the overall diarization error rate is shown to improve by 6.4 % relative.

KW - Convolutive nonnegative sparse coding

KW - Speaker diarization

KW - Speech overlap detection

UR - http://www.scopus.com/inward/record.url?scp=84878541035&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84878541035

SN - 9781622767595

T3 - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

SP - 2151

EP - 2154

BT - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

T2 - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

Y2 - 9 September 2012 through 13 September 2012

ER -

Geiger JT, Vipperla R, Bozonnet S, Evans N, Schuller B , Rigoll G. Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization. in 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 2012. S. 2151-2154. (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012).

Convolutive non-negative sparse coding and new features for speech overlap handling in speaker diarization

Abstract

Publikationsreihe

Konferenz

Andere Dateien und Links

Fingerprint

Dieses zitieren