Multimodal Dataset Normalization and Perceptual Validation for Music-Taste Correspondences
For researchers studying cross-modal correspondences between music and taste, this provides a scalable method to generate large datasets, though the approach is incremental.
The authors address the bottleneck of collecting large aligned cross-modal datasets for music-flavor research by showing that audio-flavor correlations transfer from a small experimental dataset to a large synthetic corpus, and that computational flavor targets align with human perception (Mantel r=0.45, p<0.0001).
Collecting large, aligned cross-modal datasets for music-flavor research is difficult because perceptual experiments are costly and small by design. We address this bottleneck through two complementary experiments. The first tests whether audio-flavor correlations, feature-importance rankings, and latent-factor structure transfer from an experimental soundtracks collection (257~tracks with human annotations) to a large FMA-derived corpus ($\sim$49,300 segments with synthetic labels). The second validates computational flavor targets -- derived from food chemistry via a reproducible pipeline -- against human perception in an online listener study (49~participants, 20~tracks). Results from both experiments converge: the quantitative transfer analysis confirms that cross-modal structure is preserved across supervision regimes, and the perceptual evaluation shows significant alignment between computational targets and listener ratings (permutation $p<0.0001$, Mantel $r=0.45$, Procrustes $m^2=0.51$). Together, these findings support the conclusion that sonic seasoning effects are present in synthetic FMA annotations. We release datasets and companion code to support reproducible cross-modal AI research.