Research on Innovative Microphone Array Technology for Recognition and Understanding of Acoustic Environments

2022 Fiscal Year Final Research Report

• Project/Area Number: 19H04131
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 61010:Perceptual information processing-related
• Research Institution: Waseda University (2020-2022); University of Tsukuba (2019)
• Principal Investigator: Makino Shoji 早稲田大学, 理工学術院(情報生産システム研究科・センター), 特任教授 (60396190)
• Co-Investigator(Kenkyū-buntansha): 猿渡 洋 東京大学, 大学院情報理工学系研究科, 教授 (30324974) ; 山田 武志 筑波大学, システム情報系, 准教授 (20312829)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: ブラインド音源分離 / 音響イベント検出 / 音情景解析

Outline of Final Research Achievements

In recent years, various efforts have been made under government leadership towards the realization of a highly smart society. Developing fundamental technologies for multimodal communication is also an urgent issue that needs to be addressed to solve these problems. Establishing fundamental technologies for statistical mathematics and fast signal processing to analyze and understand the sound environment, which is the core of multimodal communication, is necessary for applying it to smart security, elderly monitoring, robot audition, and other areas. This research focused on developing fundamental technologies for multimodal communication centered on the sound environment for social implementation. Specifically, we conducted research on the fundamental technologies of acoustic measurement using virtual microphones and distributed microphone arrays, as well as understanding the sound environment through deep learning.

Free Research Field

音響信号処理

Academic Significance and Societal Importance of the Research Achievements

本研究では、我々がこれまで提案してきたバーチャルマイクロホンという新概念に、音声や音響信号のパワフルな統計モデルや先進的な最適化手法を取り入れ、新しい分散型マイクロホンアレーシステムや信号処理アルゴリズムを開発したことに学術的意義がある。さらに、本研究では、マイクロホンアレー信号処理の積年の問題に立ち向かい、新しいロバストな分散型マイクロホンアレー信号処理アルゴリズムを考案し、包括的で安定な解法を開発したことに社会的意義がある。