Suppr超能文献

用于立体声重放中扩展最佳听音位置的最优指向性图案的推导。

Derivation of an optimal directivity pattern for sweet spot widening in stereo sound reproduction.

作者信息

Ródenas Josep A, Aarts Ronald M, Janssen A J E M

机构信息

Philips Research Laboratories Eindhoven, Professor Holstlaan 4, NL-5656 AA Eindhoven, The Netherlands.

出版信息

J Acoust Soc Am. 2003 Jan;113(1):267-78. doi: 10.1121/1.1527928.

Abstract

In this paper the correction of the degradation of the stereophonic illusion during sound reproduction due to off-center listening is investigated. The main idea is that the directivity pattern of a loudspeaker array should have a well-defined shape such that a good stereo reproduction is achieved in a large listening area. Therefore, a mathematical description to derive an optimal directivity pattern l(opt) that achieves sweet spot widening in a large listening area for stereophonic sound applications is described. This optimal directivity pattern is based on parametrized time/intensity trading data coming from psycho-acoustic experiments within a wide listening area. After the study, the required digital FIR filters are determined by means of a least-squares optimization method for a given stereo base setup (two pair of drivers for the loudspeaker arrays and 2.5-m distance between loudspeakers), which radiate sound in a broad range of listening positions in accordance with the derived l(opt). Informal listening tests have shown that the l(opt) worked as predicted by the theoretical simulations. They also demonstrated the correct central sound localization for speech and music for a number of listening positions. This application is referred to as "Position-Independent (PI) stereo."

摘要

本文研究了在声音重放过程中,由于偏离中心聆听而导致的立体声幻觉退化的校正问题。主要思想是扬声器阵列的指向性图案应具有明确的形状,以便在较大的聆听区域实现良好的立体声重放。因此,描述了一种数学描述,用于推导在较大聆听区域实现立体声应用中最佳聆听位置扩展的最佳指向性图案l(opt)。这种最佳指向性图案基于在较宽聆听区域内心理声学实验得到的参数化时间/强度权衡数据。研究之后,对于给定的立体声基础设置(扬声器阵列的两对驱动器以及扬声器之间2.5米的距离),通过最小二乘优化方法确定所需的数字FIR滤波器,这些滤波器根据推导的l(opt)在广泛的聆听位置辐射声音。非正式聆听测试表明,l(opt)的效果与理论模拟预测的一致。它们还证明了在多个聆听位置对语音和音乐的正确中央声音定位。这种应用被称为“位置无关(PI)立体声”。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验