Data, stimuli, and acoustic-model outputs for "Acoustic Hardness, Not Punch, Organises Perceived Heaviness in Professionally Produced Metal"

Résumé fourni par la source

This record contains the data, audio stimuli, and acoustic-model outputs underlying the article “Acoustic Hardness, Not Punch, Organises Perceived Heaviness in Professionally Produced Metal” (Herbst, Czedik-Eysenberg, Smialek, Fenton, Mynett, and Reuter), submitted to Frontiers in Psychology. Study design: Nine professional producers independently mixed one metal song. From each mix, two contrasting sections (one fast and dense, one slow and anthemic) were drawn, giving 18 listening stimuli. Seventy-nine listeners rated each stimulus on five perceptual attributes (heaviness, brightness, roughness, clarity, and punch) in an anonymous online survey. Contents: audio/section_stimuli/: the 18 rated section stimuli (the audio listeners heard). audio/full_mixes/: the nine final producer mixes (loudness-matched masters at -10.4 LUFS). data/: per-stimulus mean ratings, the de-identified participant-level ratings (n = 79), and an id-to-survey-language map. data/acoustic_model_outputs/: hardness scores and PM95 transient-loudness values at mix and stem level, plus per-stem timbral features. README.txt: a file guide and usage notes. Acoustic measurement tools: The hardness model (Czedik-Eysenberg et al., 2017, 2024) and the PM95 transient-loudness model (Fenton and Lee, 2019) are proprietary third-party tools. This record provides their outputs and the processed data tables, not the tools themselves. Participant data: The survey was anonymous and collected no direct identifiers. Free-text responses were removed before deposit, and demographic fields are retained for the reported subgroup analyses. The data are shared under the consent and ethics approval that governed the study.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.