The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Bertujuan untuk pemantauan jangka panjang perbualan pejabat harian tanpa merekod kandungan perbualan, sistem dibentangkan untuk menganggar maklumat bukan lisan akustik seperti tempoh ujaran, kekerapan ujaran dan pengambilan giliran. Sistem ini menggabungkan teknik penyetempatan bunyi berdasarkan pengagihan tenaga bunyi dengan 16 modul susunan mikrofon pembentuk rasuk yang dipasang di siling untuk mengurangkan pengaruh pantulan bunyi berbilang. Tambahan pula, pengesanan manusia menggunakan kamera medan pandangan yang luas disepadukan dengan sistem untuk anggaran pembesar suara yang lebih mantap. Sistem menganggarkan penutur untuk setiap ujaran dan mengira maklumat bukan lisan berdasarkannya. Data menganalisis penilaian yang dikumpul selama sepuluh hari bekerja 12 jam di pejabat dengan tiga pekerja yang ditugaskan menunjukkan bahawa sistem itu mempunyai 72% ketepatan pengesanan segmentasi pertuturan dan 86% ketepatan pengenalan pembesar suara apabila sebutan dikesan dengan betul. Walaupun dengan pengesanan suara palsu dan pengenalan pembesar suara yang salah dan walaupun dalam kes di mana peserta kerap membuat bising atau di mana tujuh peserta telah berkumpul bersama untuk perbincangan, susunan jumlah maklumat bukan lisan akustik yang dikira yang diucapkan oleh peserta bertepatan dengan yang berdasarkan manusia. -maklumat bukan lisan akustik berkod. Analisis berterusan dinamik komunikasi seperti penguasaan dan peranan penyertaan perbualan melalui maklumat bukan lisan akan mendedahkan dinamik sesuatu kumpulan. Sumbangan utama kajian ini adalah untuk menunjukkan kebolehlaksanaan pemantauan jangka panjang tanpa batasan aktiviti pejabat harian melalui maklumat bukan lisan akustik.
Hitomi YOKOYAMA
Graduate School of Tokyo University of Agriculture and Technology
Masano NAKAYAMA
Graduate School of Tokyo University of Agriculture and Technology
Hiroaki MURATA
Graduate School of Tokyo University of Agriculture and Technology
Kinya FUJITA
Graduate School of Tokyo University of Agriculture and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Hitomi YOKOYAMA, Masano NAKAYAMA, Hiroaki MURATA, Kinya FUJITA, "Development of Acoustic Nonverbal Information Estimation System for Unconstrained Long-Term Monitoring of Daily Office Activity" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 2, pp. 331-345, February 2019, doi: 10.1587/transinf.2018EDK0005.
Abstract: Aimed at long-term monitoring of daily office conversations without recording the conversational content, a system is presented for estimating acoustic nonverbal information such as utterance duration, utterance frequency, and turn-taking. The system combines a sound localization technique based on the sound energy distribution with 16 beam-forming microphone-array modules mounted in the ceiling for reducing the influence of multiple sound reflection. Furthermore, human detection using a wide field of view camera is integrated to the system for more robust speaker estimation. The system estimates the speaker for each utterance and calculates nonverbal information based on it. An evaluation analyzing data collected over ten 12-hour workdays in an office with three assigned workers showed that the system had 72% speech segmentation detection accuracy and 86% speaker identification accuracy when utterances were correctly detected. Even with false voice detection and incorrect speaker identification and even in cases where the participants frequently made noise or where seven participants had gathered together for a discussion, the order of the amount of calculated acoustic nonverbal information uttered by the participants coincided with that based on human-coded acoustic nonverbal information. Continuous analysis of communication dynamics such as dominance and conversation participation roles through nonverbal information will reveal the dynamics of a group. The main contribution of this study is to demonstrate the feasibility of unconstrained long-term monitoring of daily office activity through acoustic nonverbal information.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDK0005/_p
Salinan
@ARTICLE{e102-d_2_331,
author={Hitomi YOKOYAMA, Masano NAKAYAMA, Hiroaki MURATA, Kinya FUJITA, },
journal={IEICE TRANSACTIONS on Information},
title={Development of Acoustic Nonverbal Information Estimation System for Unconstrained Long-Term Monitoring of Daily Office Activity},
year={2019},
volume={E102-D},
number={2},
pages={331-345},
abstract={Aimed at long-term monitoring of daily office conversations without recording the conversational content, a system is presented for estimating acoustic nonverbal information such as utterance duration, utterance frequency, and turn-taking. The system combines a sound localization technique based on the sound energy distribution with 16 beam-forming microphone-array modules mounted in the ceiling for reducing the influence of multiple sound reflection. Furthermore, human detection using a wide field of view camera is integrated to the system for more robust speaker estimation. The system estimates the speaker for each utterance and calculates nonverbal information based on it. An evaluation analyzing data collected over ten 12-hour workdays in an office with three assigned workers showed that the system had 72% speech segmentation detection accuracy and 86% speaker identification accuracy when utterances were correctly detected. Even with false voice detection and incorrect speaker identification and even in cases where the participants frequently made noise or where seven participants had gathered together for a discussion, the order of the amount of calculated acoustic nonverbal information uttered by the participants coincided with that based on human-coded acoustic nonverbal information. Continuous analysis of communication dynamics such as dominance and conversation participation roles through nonverbal information will reveal the dynamics of a group. The main contribution of this study is to demonstrate the feasibility of unconstrained long-term monitoring of daily office activity through acoustic nonverbal information.},
keywords={},
doi={10.1587/transinf.2018EDK0005},
ISSN={1745-1361},
month={February},}
Salinan
TY - JOUR
TI - Development of Acoustic Nonverbal Information Estimation System for Unconstrained Long-Term Monitoring of Daily Office Activity
T2 - IEICE TRANSACTIONS on Information
SP - 331
EP - 345
AU - Hitomi YOKOYAMA
AU - Masano NAKAYAMA
AU - Hiroaki MURATA
AU - Kinya FUJITA
PY - 2019
DO - 10.1587/transinf.2018EDK0005
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2019
AB - Aimed at long-term monitoring of daily office conversations without recording the conversational content, a system is presented for estimating acoustic nonverbal information such as utterance duration, utterance frequency, and turn-taking. The system combines a sound localization technique based on the sound energy distribution with 16 beam-forming microphone-array modules mounted in the ceiling for reducing the influence of multiple sound reflection. Furthermore, human detection using a wide field of view camera is integrated to the system for more robust speaker estimation. The system estimates the speaker for each utterance and calculates nonverbal information based on it. An evaluation analyzing data collected over ten 12-hour workdays in an office with three assigned workers showed that the system had 72% speech segmentation detection accuracy and 86% speaker identification accuracy when utterances were correctly detected. Even with false voice detection and incorrect speaker identification and even in cases where the participants frequently made noise or where seven participants had gathered together for a discussion, the order of the amount of calculated acoustic nonverbal information uttered by the participants coincided with that based on human-coded acoustic nonverbal information. Continuous analysis of communication dynamics such as dominance and conversation participation roles through nonverbal information will reveal the dynamics of a group. The main contribution of this study is to demonstrate the feasibility of unconstrained long-term monitoring of daily office activity through acoustic nonverbal information.
ER -