senses-chant: Data and derived outputs for "Hearing and Smelling the Sacred. Sensory language in fourteenth-century Office chant across sacred settings" (paper-v2)

Research data bundle for the article "Hearing and Smelling the Sacred. Sensory language in fourteenth-century Office chant across sacred settings" by Ángel Manuel Olmos Sáez. This deposit (paper-v2) supersedes the previous version and includes all materials from the original release plus the following additions from revision round 2: - Chant-level audit of excluded rows (excluded_rows_audit/): desglose of 133 excluded odour candidates across 72 unique cantus IDs, confirming that 46/72 are already present in the retained odour field and 26 genuinely absent cases carry no validated positive- Klosterneuburg institutional sensitivity test (klosterneuburg_sensitivity.tsv): portability baseline recomputed under alternative monastic reclassification of A-KN CCl 1018- Modal assignments for the 11-chant analytical shortlist (shortlist_modal_assignments.tsv): mode field extracted from panel source CSVs with per-source breakdown- Multi-model sound classification comparison (multimodel_sound/): benchmark_80 stratified sample results for Mistral-7B, Llama-3.2-3B, Qwen2.5-7B, and Qwen3-8B- Intra-annotator consistency packet (second_reader_packet/): 120-row blind repass with Cohen-style kappa The companion code repository is archived at:https://codeberg.org/perotinus/senses-chant (tag: paper-v2, commit: 24ffbde) Bundle contents: data_derived/, data_raw/, logs/, README.md.Excluded: Git metadata, Python environments, model weights.