Multiband SMR and Inter-Channel Phase Correlation Dataset of Grammy-Nominated Tracks (1995–2026)

This repository contains the empirical datasets, statistical outputs, visualization figures, and analysis scripts accompanying the associated publication. The study investigates long-term changes in commercial stereophonic production using a corpus of 536 Grammy-nominated recordings spanning the years 1995–2026. Two complementary stereo descriptors were calculated for every recording: Side-to-Mid Ratio (SMR, dB) Inter-Channel Phase Correlation (PC) Both descriptors were computed for one broadband reference and ten frequency bands covering the audible spectrum up to the Nyquist frequency. Repository Contents The repository includes: Raw feature datasets for all analyzed recordings Annual statistical summaries Weighted Least Squares (WLS) regression results Holm–Bonferroni corrected statistical outputs Visualization figures Python analysis scripts used to compute the stereo descriptors, perform the statistical analyses, and generate baseline figures Python Analysis Pipeline The Python scripts implement the complete analysis workflow, including multiband filtering, SMR calculation, inter-channel phase correlation analysis, yearly aggregation, Weighted Least Squares regression, Holm–Bonferroni multiple-comparison correction, and export of the resulting datasets and figures. The figures generated directly by the scripts represent the default analysis output. The figures presented in the associated publication were produced separately using the exported numerical results with revised formatting for publication. Consequently, figure layouts may differ while all reported numerical values and statistical results remain identical. Copyright Notice The original FLAC audio recordings are not included because they remain protected by copyright. Only extracted numerical features, statistical summaries, visualization figures, and analysis scripts are distributed. License The datasets, figures, and analysis scripts are distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Citation If this repository contributes to academic work, please cite both the associated publication and this Zenodo repository DOI.