A DOA based speaker diarization system for real meetings

Shoko Araki; Masakiyo Fujimoto; Kentaro Ishizuka; Hiroshi Sawada; Shoji Makino

doi:10.1109/HSCMA.2008.4538680

A DOA based speaker diarization system for real meetings

Shoko Araki^*, Masakiyo Fujimoto, Kentaro Ishizuka, Hiroshi Sawada, Shoji Makino

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

26 Citations (Scopus)

Abstract

This paper presents a speaker diarization system that estimates who spoke when in a meeting. Our proposed system is realized by using a noise robust voice activity detector (VAD), a direction of arrival (DOA) estimator, and a DOA classifier. Our previous system utilized the generalized cross correlation method with the phase transform (GCC-PHAT) approach for the DOA estimation. Because the GCC-PHAT can estimate just one DOA per frame, it was difficult to handle speaker overlaps. This paper tries to deal with this issue by employing a DOA at each time-frequency slot (TFDOA), and reports how it improves diarization performance for real meetings / conversations recorded in a room with a reverberation time of 350 ms.

Original language	English
Title of host publication	2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008
Pages	29-32
Number of pages	4
DOIs	https://doi.org/10.1109/HSCMA.2008.4538680
Publication status	Published - 2008
Externally published	Yes
Event	2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008 - Trento, Italy Duration: 2008 May 6 → 2008 May 8

Publication series

Name	2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008

Conference

Conference	2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008
Country/Territory	Italy
City	Trento
Period	08/5/6 → 08/5/8

Keywords

Diarization
Direction of arrival
Voice activity detector

ASJC Scopus subject areas

Hardware and Architecture
Electrical and Electronic Engineering
Communication

Access to Document

10.1109/HSCMA.2008.4538680

Cite this

Araki, S., Fujimoto, M., Ishizuka, K., Sawada, H., & Makino, S. (2008). A DOA based speaker diarization system for real meetings. In 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008 (pp. 29-32). Article 4538680 (2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008). https://doi.org/10.1109/HSCMA.2008.4538680

A DOA based speaker diarization system for real meetings. / Araki, Shoko; Fujimoto, Masakiyo; Ishizuka, Kentaro et al.
2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008. 2008. p. 29-32 4538680 (2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Araki, S, Fujimoto, M, Ishizuka, K, Sawada, H & Makino, S 2008, A DOA based speaker diarization system for real meetings. in 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008., 4538680, 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008, pp. 29-32, 2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008, Trento, Italy, 08/5/6. https://doi.org/10.1109/HSCMA.2008.4538680

@inproceedings{e666aa11aa084e629f072326d71599cd,

title = "A DOA based speaker diarization system for real meetings",

abstract = "This paper presents a speaker diarization system that estimates who spoke when in a meeting. Our proposed system is realized by using a noise robust voice activity detector (VAD), a direction of arrival (DOA) estimator, and a DOA classifier. Our previous system utilized the generalized cross correlation method with the phase transform (GCC-PHAT) approach for the DOA estimation. Because the GCC-PHAT can estimate just one DOA per frame, it was difficult to handle speaker overlaps. This paper tries to deal with this issue by employing a DOA at each time-frequency slot (TFDOA), and reports how it improves diarization performance for real meetings / conversations recorded in a room with a reverberation time of 350 ms.",

keywords = "Diarization, Direction of arrival, Voice activity detector",

author = "Shoko Araki and Masakiyo Fujimoto and Kentaro Ishizuka and Hiroshi Sawada and Shoji Makino",

year = "2008",

doi = "10.1109/HSCMA.2008.4538680",

language = "English",

isbn = "9781424423385",

series = "2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008",

pages = "29--32",

booktitle = "2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008",

note = "2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008 ; Conference date: 06-05-2008 Through 08-05-2008",

}

TY - GEN

T1 - A DOA based speaker diarization system for real meetings

AU - Araki, Shoko

AU - Fujimoto, Masakiyo

AU - Ishizuka, Kentaro

AU - Sawada, Hiroshi

AU - Makino, Shoji

PY - 2008

Y1 - 2008

N2 - This paper presents a speaker diarization system that estimates who spoke when in a meeting. Our proposed system is realized by using a noise robust voice activity detector (VAD), a direction of arrival (DOA) estimator, and a DOA classifier. Our previous system utilized the generalized cross correlation method with the phase transform (GCC-PHAT) approach for the DOA estimation. Because the GCC-PHAT can estimate just one DOA per frame, it was difficult to handle speaker overlaps. This paper tries to deal with this issue by employing a DOA at each time-frequency slot (TFDOA), and reports how it improves diarization performance for real meetings / conversations recorded in a room with a reverberation time of 350 ms.

AB - This paper presents a speaker diarization system that estimates who spoke when in a meeting. Our proposed system is realized by using a noise robust voice activity detector (VAD), a direction of arrival (DOA) estimator, and a DOA classifier. Our previous system utilized the generalized cross correlation method with the phase transform (GCC-PHAT) approach for the DOA estimation. Because the GCC-PHAT can estimate just one DOA per frame, it was difficult to handle speaker overlaps. This paper tries to deal with this issue by employing a DOA at each time-frequency slot (TFDOA), and reports how it improves diarization performance for real meetings / conversations recorded in a room with a reverberation time of 350 ms.

KW - Diarization

KW - Direction of arrival

KW - Voice activity detector

UR - http://www.scopus.com/inward/record.url?scp=50449094778&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50449094778&partnerID=8YFLogxK

U2 - 10.1109/HSCMA.2008.4538680

DO - 10.1109/HSCMA.2008.4538680

M3 - Conference contribution

AN - SCOPUS:50449094778

SN - 9781424423385

T3 - 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008

SP - 29

EP - 32

BT - 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008

T2 - 2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008

Y2 - 6 May 2008 through 8 May 2008

ER -

A DOA based speaker diarization system for real meetings

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this