Dynamic perception process of the three dimensional spatial hearing

三维空间听觉的动态感知过程

• 批准号: 10835001
• 负责人: OZAWA Kenji
• 金额: $ 2.05万
• 依托单位: TOHOKU UNIVERSITY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 1998
• 资助国家: 日本
• 起止时间: 1998 至 1999
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-10835001/
• 关键词: Spatial hearing; Sound localization; Head movement; Head rotation; HRTF; DSP; Motion sensor; Dynamic perception

In the present study, a new spatial hearing control system, in which the head-related transfer functions (HRTF) are switched over by sensing the motion of listener's head was first developed. This system is consisted of multiple DSP chips of 50 MFLOPS in speed and realizes very short delay time of only 13 ms in switching HRTFs against the head motion. By using the above mentioned system, the psychometric measurement of the detection threshold and the difference limen (DL) of the delay time in the switching of HRTF become possible. As results, it was first showed that the detection threshold is around 50 ms and the DL suddenly increase for delay time longer than 100 ms; these finding are new, interesting and important ones. Then, a psychoacoustical experiments were conducted to examine the human ability of sound localization under two conditions of head moving. In one condition, listeners were just allowed to move their heads while in the other condition, listeners were requested to move their heads actively to locate the sound image as precisely as possible. Low-passed noise, High-passed noise as well as wide-band noise with duration from 100 ms to 3 s were used as sound stimuli. As a result, the effect of active head movement was particularly effective to reduce localization errors for the low-passed noise. Moreover, the loci of the head movements were recorded with a motion sensor. The results showed some interesting tendencies such as clear vertical motions during the localization of a sound image on the horizontal plane.

在本研究中，一个新的空间听觉控制系统，其中的头部相关的传递函数（HRTF）切换通过感知听者的头部运动的第一次开发。该系统由多个50 MFLOPS的DSP芯片组成，实现了相对于头部运动切换HRTF的仅13 ms的极短延迟时间。通过使用上述系统，可以对HRTF切换中的检测阈值和延迟时间的差阈（DL）进行心理测量。作为结果，它是第一次表明，检测阈值约为50毫秒和DL突然增加的延迟时间长于100毫秒，这些发现是新的，有趣的和重要的。在此基础上，通过心理声学实验研究了人在两种头部运动情况下的声音定位能力。在一种情况下，听众只被允许移动他们的头部，而在另一种情况下，听众被要求主动移动他们的头部，以尽可能精确地定位声像。分别用低通噪声、高通噪声和持续时间为100 ms ~ 3 s的宽带噪声作为声刺激。因此，主动头部运动的效果对于降低低通噪声的定位误差特别有效。此外，用运动传感器记录头部运动的轨迹。结果表明，一些有趣的趋势，如明确的垂直运动过程中的本地化的声像在水平面上。

项目成果

Daisuke Kimura: "The detection threshold of the delay of a sound image against head movement"Proc. Of Tohoku-Section Joint Convention Record of Institutes of Electrical and Information Engineer. 2A8 (1999)

Daisuke Kimura：“头部运动声像延迟的检测阈值”Proc。

Shouichi Takane: "Proposal of a new auditory display based on virtual sphere model"Proc. Cong. Of Acoust. Soc. Jpn.. 501-502 (1998)

Shouichi Takane：“基于虚拟球体模型的新听觉显示的提议”Proc。

木村大助: "頭部運動による音像移動の遅延検知限に関する考察"電気関係学会東北支部大会講演論文集. 2A8 (1999)

Daisuke Kimura：“头部运动引起的声像运动延迟检测极限的研究”日本电气工程师学会东北分会会议记录2A8（1999）。

木村大助: "頭部運動に対する音像移動遅延の検知限と弁別域"日本音響学会講演論文集. 1-10-18 (2000)

Daisuke Kimura：“头部运动的声像运动延迟的检测极限和辨别范围”日本声学学会会议记录1-10-18（2000）。