基于双耳时间差的客观指标对Ambisonics回放系统的性能分析

发布时间：2018-01-06 15:32

本文关键词：基于双耳时间差的客观指标对Ambisonics回放系统的性能分析　出处：《南京大学学报(自然科学)》2017年06期 　论文类型：期刊论文

【摘要】：在基于扬声器阵列的声回放技术中,相比于波场合成(Wave field synthesis,WFS)和矢量基幅度平移技术,以球谐分解为理论基础的Ambisonics回放系统拥有编解码相互独立以及可扩展等优势.基本的Ambisonics系统通过采集和回放一阶声场方位信息:全指向性(W)和双指向性(X,Y,Z)成分,即所谓的B-format重构声场.HOA(Higher order Ambisonics)基于声场球谐函数分解将B-format以更高空间分辨率进行了扩展.已有很多学者关注HOA的回放精度和使用限制,但综合考虑人头散射效应和人耳听觉特性效应的研究还十分缺乏.从低频人耳定位的重要因素——双耳时间差(Interaural time difference,ITD)的角度评价了不同阶数Ambisonis系统的最佳听音区域.将水平面各方向入射平面波编码为HOA分量,在此基础上计算了模拟人头(刚性球)在环形阵列内部移动时二维水平面的ITD波动,并通过ITD阈值确定最佳听音区域边界.仿真结果表明基于ITD的客观评价指标可以较好地体现不同阶数Ambisonics系统的声场回放性能:4阶Ambisoics系统能够使最佳听音区域达到20cm×14cm,而1,2阶系统在中心区域尚不能实现精确回放.因此,高阶Ambisoics系统拥有更好的声源定位性能.
[Abstract]:Based on the sound playback technology of loudspeaker array, compared to Yu Bo field (Wave field synthesis, WFS synthesis) and vector base amplitude panning technique, with spherical harmonic decomposition Ambisonics playback system based codec has independent and extensible advantages. The basic Ambisonics system through the acquisition and playback of one order acoustic orientation information: the whole point (W) and double point (X, Y, Z), called B-format (Higher order Ambisonics.HOA sound field reconstruction) field of the spherical harmonic decomposition of B-format at higher spatial resolution for the expansion. Based on playback precision and limits of many scholars pay attention to the HOA, but considering the study on the head scattering effect and the characteristics of human hearing effect is still lacking. The important factors from the positioning ear: the low frequency interaural time difference (Interaural time, difference, ITD) to evaluate the different order The best listening area number Ambisonis. The horizontal surface of each direction of the incident plane wave encoding for the HOA component, were calculated based on the simulation of head (rigid ball) in the annular array inside the Mobile two-dimensional horizontal plane ITD wave, and through the ITD threshold to determine the best listening region boundary. The simulation results show that the sound playback performance objective the evaluation index based on ITD can better reflect the different order of Ambisonics system: 4 order Ambisoics system can make the best listening area to 20cm * 14cm, and the 1,2 order system to achieve accurate playback in the central region is not. Therefore, the sound source localization performance of high-order Ambisoics system has better.

【作者单位】：南京大学声学研究所近代声学教育部重点实验室;
【基金】：国家自然科学基金(11374156,11474163)
【分类号】：TN912
【正文快照】： 空间声回放技术主要分为两类:一类是基于感知的虚拟声场景重建技术如基于矢量的幅度平移技术(Vector base amplitude panning,VBAP)[1-3],另一类是物理声场重建技术,如波场合成(Wave field synthesis,WFS)[4-6].前者考虑人耳的感知特性,实现相对简单;后者则追求对原始声场的精

【相似文献】