A study of acoustic cross-reality based on selective listening point audio

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K12193
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61060:Kansei informatics-related
• Research Institution: Meijo University
• Principal Investigator: NISHINO Takanori 名城大学, 都市情報学部, 教授 (40329769)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 音響クロスリアリティ / 立体音響 / 頭部伝達関数 / 深層学習応用 / 快音・不快音

Outline of Final Research Achievements

This study investigated a realization of acoustic cross reality such as virtual reality, augmented reality, and mixed reality, with the subject of three-dimensional acoustic reproduction of the sound environment desired by the user and provided to the user. From the viewpoint of providing a comfortable sound environment, a large-scale survey of discomfort environmental sounds was conducted, and a method of converting one's recorded voice to one's own heard voice was also proposed. In addition, a sound source localization model based on deep learning was examined. The goal of this sound localization model will achieve its performance to the same degree as humans. The experiments using different acoustic signals for training and evaluation were conducted and revealed its performance and some problems.

Free Research Field

音響工学

Academic Significance and Societal Importance of the Research Achievements

本課題は、立体音響を活用した音響情報提示の基礎検討を行ったものである。ユーザに違和感を抱かせないような音響提示を行うための検討からは、世代間の差を明らかにするとともに、既存の音声処理技術を用いても有効な音声加工が可能であることを確認した。これらの結果から、違和感が少ない音響提示が可能であることが示唆された。また、深層学習応用から、人間と同様の音源定位を行うために必要なモデルや学習方法についての課題について整理することができた。本課題の成果は、仮想現実技術やロボット技術において活用できると考える。