Outdoor Scene Understanding for Bird Song Analysis

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K00294
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Nakadai Kazuhiro 東京工業大学, 工学院, 特任教授 (70436715)
• Co-Investigator(Kenkyū-buntansha): 鈴木 麗璽 名古屋大学, 情報学研究科, 准教授 (20362296) ; 小島 諒介 京都大学, 医学研究科, 特定助教 (70807651)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: ロボット聴覚 / 人工知能 / 音響信号処理 / 行動学 / 情報システム

Outline of Final Research Achievements

From the singing voice sound signals of wild birds recorded by multiple microphone arrays, we developed outdoor sound environment understanding technology that extracts structured information on what, when and where of bird singing voice events, and that estimates the relationship between wild birds, by integrating robot audition and machine learning technologies. In addition, we have built an outdoor sound environment understanding system for bird song analysis that is easy to use even for a non-expert, and reduce the burden of singing voice analysis work on wild birds that has been performed manually, which resulted in contributing to the field of animal behavioral sciences and bioacoustics.

Free Research Field

ロボット聴覚，音環境理解

Academic Significance and Societal Importance of the Research Achievements

本研究は，従来のロボット聴覚技術と野鳥歌解析技術の欠点を相補的に補うことで，人手を介せずに野鳥の歌声解析に適用可能な屋外音環境理解の枠組みを構築する点で学術的意義が大きい．また，，基礎的な研究課題への取り組みだけではなく，野鳥行動解析の専門家を分担者に招くことで，ユーザ視点での技術課題抽出を行い，必ずしも技術の専門家でない野鳥の行動解析の研究者であっても，簡単に操作できる自動歌声解析システムを構築することで，実作業の負担軽減を狙う点で社会的な意義が高い．