• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Conversation analysis for detecting age-related decline and its evaluation

Research Project

Project/Area Number 19K04934
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 25020:Safety engineering-related
Research InstitutionOsaka Institute of Technology

Principal Investigator

Wakita Yumi  大阪工業大学, ロボティクス&デザイン工学部, 教授 (10590359)

Co-Investigator(Kenkyū-buntansha) 中藤 良久  九州工業大学, 大学院工学研究院, 教授 (10599955)
松田 千登勢  摂南大学, 看護学部, 教授 (70285328)
Project Period (FY) 2019-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000)
Fiscal Year 2021: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2020: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2019: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Keywords加齢による衰え推定 / 日常会話 / 頭部動作の変化量 / 話題の広がり度 / マイクロホン距離 / 会話理解度 / オンライン日常会話 / 高齢者サービス住宅 / 応答発声の音響特性 / 高齢者会話特徴 / 会話理解度推定 / 応答発声 / ピッチ抽出 / オンライン会話データ / 衰え度察知システム / 日常会話解析 / 高齢者会話特性 / 基本周波数の時間変化 / トピックシフト特性 / 理解力衰え推定 / 目標会話識別 / 応答時の非言語情報 / 時間変化成分 / 目標音声識別 / パワー時間変化分布 / 応答しぐさ / 衰え推定 / 非言語情報
Outline of Research at the Start

高齢者交通事故の原因は「運転操作不適」であり、高齢者に「衰え」の自覚がなく自らを過信している問題が指摘されている。日常生活の中でさりげなく衰えを察知し本人に警鐘を促すシステムを実現し、事故を未然に防ぎたい。我々は、独自の高齢者日常会話データから、会話の基本周波数の分布が、会話を聞いた人が老化に気付くしくみと関連が深いこと発見し、衰え察知モデルを示した。しかし本データは衰えの経時変化を含んでいないため、提案モデルは、高齢者らしさは判断できても「加齢による衰え」を察知できるかは明確ではない。本研究では、施設にて高齢者の会話を長期に観察することで提案モデルを改良し、衰え察知モデルを確立する。

Outline of Final Research Achievements

1. Development of daily conversation database among elderly people: 22 conversations among elderly people in nursing homes and serviced residences who need daily living assistance. 10 transcriptions of the conversations were also constructed. 15 of the 22 conversations were online conversations.
2.The effectiveness of dynamic features of head motion on speakers and Topic-shift characteristics in daily conversation for age-related decline estimation: We confirmed that the amount of change in head motion during conversation tended to differ between the elderly who received daily living assistance and the elderly who lived independently. Also we confirmed that the Topic-shift characteristics in daily conversation differ between younger and older adults, and can be distinguished in 90.5% of cases between older and younger adults. These results suggests that adding this information to conventional F0 and SPL information for each uttacance will improve the accuracy of estimating decline.

Academic Significance and Societal Importance of the Research Achievements

【学術的意義】相手の発話の理解度が応答時発声のピッチやパワーの継時変化、会話者の頭部の動きの変化量、話題の広がりなどで推定できたことと、本推定が、年齢や会話手段(対面かオンラインか)に寄らず精度よくできることを確認できたことは、会話音声分析分野にて意義があると思われる.
【社会的意義】高齢者における加齢による衰え度合の推定を行うことを目的とする.認知症に至る前の段階での理解力の低下が日常会話の発話の音響的特徴として現れると仮定し、理解力の僅かな低下を推定可能な技術は、高齢者にいち早く免許返納の時期を促すことができ、高齢ドライバーの事故の減少に貢献すると考える,

Report

(6 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • 2020 Research-status Report
  • 2019 Research-status Report
  • Research Products

    (23 results)

All 2024 2022 2021 2020 2019

All Journal Article (7 results) (of which Int'l Joint Research: 2 results,  Peer Reviewed: 7 results,  Open Access: 2 results) Presentation (16 results) (of which Int'l Joint Research: 15 results)

  • [Journal Article] Estimation of Online Lecture Quality Using Fundamental Frequency Characteristics Extracted from Student utterances2022

    • Author(s)
      Tomoyuki Takata, Yumi Wakita
    • Journal Title

      Learning and Collaboration Technologies. Designing the Learner and Teacher Experience

      Volume: 13328 Pages: 304-312

    • DOI

      10.1007/978-3-031-05657-4_22

    • ISBN
      9783031056567, 9783031056574
    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Age Estimation Function Using End-of Sentence Expression for Conversation Support System Based on Associative Words Board2022

    • Author(s)
      Hikaru Nishida, Yotaro Iida, Yumi Wakita
    • Journal Title

      International Symposium on Distributed Computing and Artificial Intelligence

      Volume: 583 Pages: 240-248

    • DOI

      10.1007/978-3-031-20859-1_24

    • ISBN
      9783031208584, 9783031208591
    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Machine learning model for discrimination of mild dementia patients using acoustic features2022

    • Author(s)
      Nishikawa Kazu、Akihiro Kuwahara、Hirakawa Rin、Kawano Hideaki、Nakatoh Yoshihisa
    • Journal Title

      Cognitive Robotics

      Volume: 2 Pages: 21-29

    • DOI

      10.1016/j.cogr.2021.12.003

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Topic-Shift Characteristics of Japanese Casual Conversations Between Elderlies and Between Youths2021

    • Author(s)
      Iida Yotaro, Yumi Wakita
    • Journal Title

      Lecture Notes in Computer Science

      Volume: 12787 Pages: 418-427

    • DOI

      10.1007/978-3-030-78111-8_28

    • ISBN
      9783030781101, 9783030781118
    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Identification of target speech utterances from real lounge conversation2020

    • Author(s)
      Naoto Kosaka, Yumi Wakita
    • Journal Title

      Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management. Human Communication, Organization and Work

      Volume: 12199 Pages: 5263-5263

    • DOI

      10.1007/978-3-030-49907-5_4

    • ISBN
      9783030499068, 9783030499075
    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] Age-dependent degradation estimation focusing on dynamic features of head motion in daily conversation2020

    • Author(s)
      Natsumi Kana、Wakita Yumi、Matsuda Chitose
    • Journal Title

      2020 IEEE 9th Global Conference on Consumer Electronics (GCCE)

      Volume: - Pages: 373375-373375

    • DOI

      10.1109/gcce50665.2020.9292020

    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] Estimating Age-Dependent Degradation using Nonverbal Feature Analysis of Daily Conversation2019

    • Author(s)
      Kana Natsumi, Yumi Wakita, Yoshihisa Nakatoh
    • Journal Title

      Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management. Healthcare Applications

      Volume: 11582 Pages: 222-231

    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Presentation] Acoustic Characteristics of utterances in Online Conversations in Serviced Elderly Residences (査読前仮)2024

    • Author(s)
      Atsuteru Iida, Chitose Matsuda, Yumi Wakita
    • Organizer
      2024 IEEE 13th Global Conference on Consumer Electronics
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Estimation of Online Lecture Quality Using Fundamental Frequency Characteristics Extracted from Student utterances2022

    • Author(s)
      Tomoyuki Takata, Yumi Wakita
    • Organizer
      International Conference on Human-Computer Interaction 2022
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Age Estimation Function Using End-of Sentence Expression for Conversation Support System Based on Associative Words Board2022

    • Author(s)
      Hikaru Nishida, Yotaro Iida, Yumi Wakita
    • Organizer
      International Symposium on Distributed Computing and Artificial Intelligence
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Estimation of online lecture quality using fundamental frequency characteristics extracted from student utterances2022

    • Author(s)
      Tomoyuki Takada, Yumi Wakita
    • Organizer
      24RD INTERNATIONAL CONFERENCE ON HUMAN-COMPUTER INTERACTION
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] System of Predicting Dementia Using Transformer Based Ensemble Learning2022

    • Author(s)
      Kazu Nishikawa, Rin Hirakawa, Hideaki Kawano, Yoshihisa Nakatoh
    • Organizer
      IEEE International Conference on Consumer Electronics
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Analysis of Prosodic Features of Dementia Speech for Machine Learning2022

    • Author(s)
      Kazu Nishikawa, Rin Hirakawa, Hideaki Kawano and Yoshihisa Nakatoh
    • Organizer
      5th International Conference on Inventive Computation Technologies
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Topic-Shift Characteristics of Japanese Casual Conversations Between Elderlies and Between Youths2021

    • Author(s)
      Iida Yotaro, Yumi Wakita
    • Organizer
      23RD INTERNATIONAL CONFERENCE ON HUMAN-COMPUTER INTERACTION
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] 会話支援システムのための簡易な目標発声特定方法2021

    • Author(s)
      小坂直人 脇田由実
    • Organizer
      2021年度日本音響学会秋季研究発表会
    • Related Report
      2021 Research-status Report
  • [Presentation] Detecting System Alzheimer's Dementia by 1d CNN-LSTM in Japanese Speech2021

    • Author(s)
      Kazu Nishikawa, Rin Hirakawa, Hideaki Kawano, Kenichi Nakashi, Yoshihisa Nakatoh
    • Organizer
      Proc. of International Conference on Consumer Electronics (ICCE2021)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Identification of target speech utterances from real lounge conversation2020

    • Author(s)
      Kosaka Naoto, Yumi Wakita
    • Organizer
      22nd INTERNATIONAL CONFERENCE ON HUMAN-COMPUTER INTERACTION
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Evaluating target utterance identification method using practical free conversation2020

    • Author(s)
      Kosaka Naoto, Yumi Wakita
    • Organizer
      2nd IEEE International Conference on Artificial Intelligence in Engineering and Technology
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Effectiveness of dynamic image features for estimating daily conversation atmosphere2020

    • Author(s)
      Kana Natsumi, Yumi Wakita, Chitose Matsuda
    • Organizer
      2nd IEEE International Conference on Artificial Intelligence in Engineering and Technology
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Age-dependent degradation estimation focusing on dynamic features of head motion in daily conversation2020

    • Author(s)
      Kana Natsumi, Yumi Wakita, Chitose Matsuda
    • Organizer
      2020 IEEE 9th Global Conference on Consumer Electronics (GCCE)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Effective Speech Features for Distinguishing Mild Dementia Patients from Healthy Person2020

    • Author(s)
      Kazu Nishikawa, Rin Hirakawa, Hideaki Kawano, Kenichi Nakashi, Yoshihisa Nakatoh
    • Organizer
      Proc. of 2nd International Conference on Human Interaction and Emerging Technologies (IHIET 2020)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Estimating Age-Dependent Degradation using Nonverbal Feature Analysis of Daily Conversation2019

    • Author(s)
      Kana Natsumi, Yumi Wakita, Yoshihisa Nakatoh
    • Organizer
      International Conference on Human-Computer Interaction 2019
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] A Simple Identification Method for Differentiating between Ambient and Target Speech2019

    • Author(s)
      Naoto Kosaka, Kohei Kido, Yumi Wakita
    • Organizer
      2019 IEEE 8th Global Conference on Consumer Electronics
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2019-04-18   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi