Associated dataset for "Instrumental Evaluation of Sensor Self-Noise in Binaural Rendering of Spherical Microphone Array Signals"

The conducted instrumental evaluation utilizes the Real-Time Spherical Microphone Renderer (ReTiSAR) for binaural reproduction in Python. The at that time employed code state should be used in order to exactly reproduce the rendering results in this data set. The frozen code state for this data set is available at: https://github.com/AppliedAcousticsChalmers/ReTiSAR/releases/tag/v2020.FA Download the rendering pipeline and follow the setup instructions! Use the here included Conda environment file when setting up the Python environment. In this way, you will obtain exactly the same Python setup as utilized in the instrumental evaluation in the publication: conda env create --file ReTiSAR_environment_freeze.yml source activate ReTiSAR_FA_freeze Directory "SMA sampling grids": Visualization of spatial arrangement (like Figure 4) for all investigated spherical microphone array rendering configurations (Table 1) Shell script "record_snr.sh": Record the input and output signals of the rendering pipeline for sound field (target / wanted) and self-noise (unwanted) components for all configurations at multiple head orientations All captured signals are contained in the "SNR" directory Matlab script "calculate_snr.m": Visualize the raw captured input and output signals (like Figure 1 for all configurations) Visualize the resulting signal-to-noise ratio (like Figure 2 for all configurations) Visualize the comparison of the resulting signal-to-noise ratio of all configurations (Figure 3, also for the resulting SNR from signals with A-weighting) All generated plots are contained in the "SNR" directory Shell script "record_noise.sh": Record the calibration and noise signals of the mh acoustic Eigenmike 32 spherical microphone array in the anechoic chamber at Chalmers University of Technology (Appendix) All captured signals are contained in the "EM32 measurements" directory Pictures of the measurement setup are contained in the "Pictures" subdirectory Matlab script "calculate_EM32_noise_levels.m": Determine the resulting target signal sensitivity and equivalent input noise levels for the investigated pre-amplification gains (Table 2) Visualize the statistical distribution of the individual raw and weighted SMA channels (like Figure 6 for all configurations) Visualize the spatial distribution of the individual raw and weighted SMA channels for all configurations Visualize the smoothed and averaged magnitude spectra of the individual raw and weighted SMA channels (like Figure 5 for all configurations)

本研究开展的仪器评测工作，采用Python实现的实时球面麦克风渲染器（Real-Time Spherical Microphone Renderer, ReTiSAR）完成双耳声重放任务。需使用本数据集对应的固化代码状态，方可精准复现本数据集的渲染结果。本数据集的固化代码版本可从以下地址获取：https://github.com/AppliedAcousticsChalmers/ReTiSAR/releases/tag/v2020.FA。请下载该渲染管线并按照配置说明完成环境搭建：配置Python环境时，请使用本数据集附带的Conda环境配置文件。通过以下命令即可得到与本研究仪器评测中完全一致的Python运行环境：
conda env create --file ReTiSAR_environment_freeze.yml
source activate ReTiSAR_FA_freeze

"SMA采样网格"目录：包含所有待评估的球面麦克风阵列（Spherical Microphone Array, SMA）渲染配置（见表1）的空间排布可视化结果（对应论文图4）。
Shell脚本"record_snr.sh"：针对所有配置，在多种头部朝向场景下，分别录制声场目标信号（有效信号）与自噪声信号（干扰信号）的渲染管线输入输出信号。所有采集到的信号均存储于"SNR"目录中。
Matlab脚本"calculate_snr.m"可实现如下功能：
1. 可视化所有配置的原始采集输入与输出信号（对应论文图1）；
2. 可视化各配置的最终信噪比结果（对应论文图2）；
3. 可视化所有配置的信噪比对比结果（对应论文图3，同时包含经A计权后的信号信噪比结果）。
所有生成的图表均存储于"SNR"目录中。
Shell脚本"record_noise.sh"：用于在查尔姆斯理工大学的消声室内，录制mh acoustic Eigenmike 32球面麦克风阵列的校准信号与噪声信号（详见附录）。所有采集到的信号均存储于"EM32 measurements"目录中。
本次测量装置的实拍照片存储于"Pictures"子目录中。
Matlab脚本"calculate_EM32_noise_levels.m"可实现如下功能：
1. 计算待评估的前置放大增益对应的目标信号灵敏度与等效输入噪声水平（见表2）；
2. 可视化各配置中单条原始与加权SMA通道的统计分布（对应论文图6）；
3. 可视化所有配置中单条原始与加权SMA通道的空间分布；
4. 可视化所有配置中单条原始与加权SMA通道的平滑平均幅度谱（对应论文图5）。