Audio head pose estimation using the direct to reverberant speech ratio

Mark Barnard; Wenwu Wang; Josef Kittler

doi:10.1109/ICASSP.2013.6639234

Audio head pose estimation using the direct to reverberant speech ratio

Mark Barnard, Wenwu Wang, Josef Kittler

Research output: Contribution to conference › Paper › peer-review

Abstract

Head pose is an important cue in many applications such as, speech recognition and face recognition. Most approaches to head pose estimation to date have used visual information to model and recognise a subject's head in different configurations. These approaches have a number of limitations such as, inability to cope with occlusions, changes in the appearance of the head, and low resolution images. We present here a novel method for determining coarse head pose orientation purely from audio information, exploiting the direct to reverberant speech energy ratio (DRR) within a highly reverberant meeting room environment. Our hypothesis is that a speaker facing towards a microphone will have a higher DRR and a speaker facing away from the microphone will have a lower DRR. This hypothesis is confirmed by experiments conducted on the publicly available AV16.3 database. © 2013 IEEE.

Original language	English
DOIs	https://doi.org/10.1109/ICASSP.2013.6639234
Publication status	Published - May 2013
Event	38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) - Vancouver, Canada Duration: 26 May 2013 → 31 May 2013

Conference

Conference	38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
Period	26/05/13 → 31/05/13

Bibliographical note

Note: Published in: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ : Institute of Electrical and Electronics Engineers. pp. 8056-8060. ISSN (print) 1520-6149 ISSN (online) 2379-190X ISBN 9781479903566

Keywords

Computer science and informatics

Access to Document

10.1109/ICASSP.2013.6639234

http://dx.doi.org/10.1109/ICASSP.2013.6639234

1 Conference contribution

Audio head pose estimation using the direct to reverberant speech ratio
Barnard, M., Wang, W. & Kittler, J., May 2013, Published in: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ : Institute of Electrical and Electronics Engineers. pp. 8056-8060. ISSN (print) 1520-6149 ISSN (online) 2379-190X ISBN 9781479903566.
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Cite this

@conference{109ab0e7e4df45c5a5d880a569969d20,

title = "Audio head pose estimation using the direct to reverberant speech ratio",

abstract = "Head pose is an important cue in many applications such as, speech recognition and face recognition. Most approaches to head pose estimation to date have used visual information to model and recognise a subject's head in different configurations. These approaches have a number of limitations such as, inability to cope with occlusions, changes in the appearance of the head, and low resolution images. We present here a novel method for determining coarse head pose orientation purely from audio information, exploiting the direct to reverberant speech energy ratio (DRR) within a highly reverberant meeting room environment. Our hypothesis is that a speaker facing towards a microphone will have a higher DRR and a speaker facing away from the microphone will have a lower DRR. This hypothesis is confirmed by experiments conducted on the publicly available AV16.3 database. {\textcopyright} 2013 IEEE.",

keywords = "Computer science and informatics",

author = "Mark Barnard and Wenwu Wang and Josef Kittler",

note = "Note: Published in: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ : Institute of Electrical and Electronics Engineers. pp. 8056-8060. ISSN (print) 1520-6149 ISSN (online) 2379-190X ISBN 9781479903566; 38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) ; Conference date: 26-05-2013 Through 31-05-2013",

year = "2013",

month = may,

doi = "10.1109/ICASSP.2013.6639234",

language = "English",

}

TY - CONF

T1 - Audio head pose estimation using the direct to reverberant speech ratio

AU - Barnard, Mark

AU - Wang, Wenwu

AU - Kittler, Josef

N1 - Note: Published in: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ : Institute of Electrical and Electronics Engineers. pp. 8056-8060. ISSN (print) 1520-6149 ISSN (online) 2379-190X ISBN 9781479903566

PY - 2013/5

Y1 - 2013/5

N2 - Head pose is an important cue in many applications such as, speech recognition and face recognition. Most approaches to head pose estimation to date have used visual information to model and recognise a subject's head in different configurations. These approaches have a number of limitations such as, inability to cope with occlusions, changes in the appearance of the head, and low resolution images. We present here a novel method for determining coarse head pose orientation purely from audio information, exploiting the direct to reverberant speech energy ratio (DRR) within a highly reverberant meeting room environment. Our hypothesis is that a speaker facing towards a microphone will have a higher DRR and a speaker facing away from the microphone will have a lower DRR. This hypothesis is confirmed by experiments conducted on the publicly available AV16.3 database. © 2013 IEEE.

AB - Head pose is an important cue in many applications such as, speech recognition and face recognition. Most approaches to head pose estimation to date have used visual information to model and recognise a subject's head in different configurations. These approaches have a number of limitations such as, inability to cope with occlusions, changes in the appearance of the head, and low resolution images. We present here a novel method for determining coarse head pose orientation purely from audio information, exploiting the direct to reverberant speech energy ratio (DRR) within a highly reverberant meeting room environment. Our hypothesis is that a speaker facing towards a microphone will have a higher DRR and a speaker facing away from the microphone will have a lower DRR. This hypothesis is confirmed by experiments conducted on the publicly available AV16.3 database. © 2013 IEEE.

KW - Computer science and informatics

U2 - 10.1109/ICASSP.2013.6639234

DO - 10.1109/ICASSP.2013.6639234

M3 - Paper

T2 - 38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

Y2 - 26 May 2013 through 31 May 2013

ER -

Audio head pose estimation using the direct to reverberant speech ratio

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Research output

Audio head pose estimation using the direct to reverberant speech ratio

Cite this