Psychoacoustic analysis and spatial coding of microphone array signals

麦克风阵列信号的心理声学分析和空间编码

• 批准号: 192387431
• 负责人: Professor Dr.-Ing. Karlheinz Brandenburg
• 金额: --
• 依托单位: Institut für Medientechnik
• 依托单位国家: 德国
• 项目类别: Research Units
• 财政年份: 2011
• 资助国家: 德国
• 起止时间: 2010-12-31 至 2013-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/192387431?language=en
• 关键词: Psychoacoustic; analysis; spatial; coding; microphone

Microphone arrays used for measuring and recording of spatial audio signals are never perfect and this measurement error can affect the perceived quality for instance for sound field reproduction. Examples of such errors are microphone noise, positioning inaccuracies of the array, spatial aliasing, non-cardioid microphone characteristics, etc. Until now those errors have been analyzed and optimized with respect to acoustical measurements only. This subproject will investigate psychoacoustic methods for the evaluation of spatial sampling and sensor errors and use these methods for creating a model for the estimation of the perceived quality. This model will be used for the improvement of microphone arrays and also the efficient storage of spatial sound fields for the further use in measurement and auralisation. For the evaluation of the spatial quality of sampled audio a proper spatial algorithm to do an averaging adjusting the spatial, temporal and spectral resolution to the resolution of the human auditory system will be investigated. Finally a file format for the efficient storage of spatial audio signals recorded with microphone arrays will be defined and an encoder for the raw measurement data will be implemented.For the correlation between measurement errors from microphone arrays and psychoacoustic models perceptual audio measurement methods as well as listening tests will be performed. The subproject will incorporate the knowledge about perception thresholds obtained in subproject P2, the acoustic attributes of P7 and the accuracy studies of microphone arrays and measurement conducted in P3 and P8.

用于测量和记录空间音频信号的麦克风阵列从来都不是完美的，这种测量误差可能会影响感知质量，例如声场再现的质量。此类误差的示例包括麦克风噪声、阵列定位不准确、空间混叠、非心形麦克风特性等。到目前为止，这些误差仅针对声学测量进行了分析和优化。该子项目将研究用于评估空间采样和传感器误差的心理声学方法，并使用这些方法创建用于估计感知质量的模型。该模型将用于改进麦克风阵列以及空间声场的有效存储，以便进一步用于测量和可听化。为了评估采样音频的空间质量，将研究适当的空间算法来进行平均，将空间、时间和频谱分辨率调整为人类听觉系统的分辨率。最后，将定义一种用于有效存储用麦克风阵列记录的空间音频信号的文件格式，并实现原始测量数据的编码器。对于麦克风阵列的测量误差和心理声学模型的感知音频测量方法以及将进行听力测试。该子项目将纳入子项目 P2 中获得的感知阈值知识、P7 的声学属性以及 P3 和 P8 中进行的麦克风阵列和测量的准确性研究。