音を聞き分ける聴覚ロボットの開発

开发能够区分声音的听觉机器人

• 批准号: 08750564
• 负责人: 黄 捷
• 金额: $ 0.58万
• 依托单位: The Institute of Physical and Chemical Research
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Encouragement of Young Scientists (A)
• 财政年份: 1996
• 资助国家: 日本
• 起止时间: 1996 至 无数据
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-08750564/
• 关键词: 聴覚ロボット; カクテルパーティ効果; 先行効果; 音源定位

われわれの生活環境では常にノイズと反響音が存在する。雑踏な環境の中で特定の音だけに注目し、ほかから聞き分ける現象をカクテルパーティ効果といい、反響環境においても反響音に影響されず、ただしく音源の位置を定位できる現象を先行効果という。人間と同じ環境で自律的に動き回るロボットを開発するにあたっては、人間と同程度の音源定位、分離能力を必要とする。音源定位システムは音源定位手掛かり処理部と抑制信号生成部からなる。抑制信号生成部は予測される反響音のレベルによって音源定位を抑制する機能をもつ。反響音の予測はインパルス応答の遅延特性と減衰特性を一般化したインパルス応答のパターンを用いる。先行効果と同じように本システムは反響音の少ない音の立ち上がり(ONSET)を検出し、音源定位を行う。反響音のレベルを先行効果のモデルによって予測できるため、ONSETの検出条件を自動的に環境に合わせることができる。音源分離システムは音を知覚の基本要素である時間周波数平面上の連続的な周波数成分で表現し、分離統合することによって違う音源から来る音の分離を行う。周波数成分の分離統合にあたっては認知的な要因を用いると同時に各成分の音源方向を先行効果によって定位し、音源の方向による分離統合を行う。先行効果とカクテルパーティ効果の実現によって、ロボットは自分の所有者などの位置を音声から認識できるようになる。また、雑踏な環境においても、命令を聴き、任務を遂行することができる。音信号の特徴として暗闇の中や物陰からもコミュニケーションできる。

The living environment is often affected by noise. In the middle of the environment, the specific sound is not noticed, the sound is not heard, the phenomenon is not affected, the environment is affected, the sound source is located, the phenomenon is affected first The ability to locate and separate sound sources in the same environment is necessary. The sound source positioning system includes a processing unit and a suppression signal generation unit. The suppression signal generation unit has the function of predicting the echo sound and suppressing the sound source localization. A generalization of the delay characteristics and attenuation characteristics of the acoustic response prediction The first result is the same as the first result. The first result is the same as the first result. The second result is the same as the first result. The third result is the same as the first result. The fourth result is that the first result is the same as the second result. The third result is that the first result is the same as the first result. The fourth result is that the first result is the same as the second result. The fourth result is that the first result is the same as the first result. The fourth result is that the first result is the same as the second result. The fourth result is that the first result is the second result. The fourth result is that the second result is the same as the third result. The fourth result is that the first result is the second result. The fourth result is that the first result is the second result is the same as the third result. The fourth result is that the first result is the second result is the first result is the second result. The fourth result is that the first result is the second result is the second result is the first result is the second result is the second result. The fourth result is that the first result is the second result is the The response time is the first time that the response time is determined, and the response time is determined. The basic elements of sound source separation and sound detection are the frequency components of continuous waves on the time frequency plane, and the separation and integration of sound source separation and sound detection. The separation and integration of the frequency components is based on the recognition of the main causes, and the sound source direction of each component is based on the prior effect, positioning, and separation and integration of the sound source direction. The first result is the realization of the result. The second result is the owner's position. The third result is the sound recognition. The environment, the command, the responsibility. The characteristics of the acoustic signal are as follows: