A study on the protection of speech privacy

2014 Fiscal Year Final Research Report

• Project/Area Number: 24500226
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Ryukoku University
• Principal Investigator: KATAOKA Akitoshi 龍谷大学, 理工学部, 教授 (20528682)
• Research Collaborator: YASUEDA Kazuya 龍谷大学, 大学院理工学研究科
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 多点制御法 / スピーチプライバシー / マスキング / 信号分解法 / 環境音

Outline of Final Research Achievements

This study describes a method of sound reproduction in the limited area. It achieves high--quality speech in the target direction. On the other hand, there is significant degradation in the outside of the target direction. This method uses two schemes; signal decomposition method and the interference method of the outside of target direction by using environmental sounds. Speech is decomposed into two sub-signals using signal decomposition method, and is produced by the multi point control method. Furthermore, we discuss configuration of multi-point control filter for reproduction of environmental sound in the outside of target direction. Subjective tests show that the proposed method gives the good performance compared with the conventional method. The proposed method is possibility to protect the privacy of the contents of the conversation.

Free Research Field

音声・音響信号処理