Three pilot source-treatises feeding the meta-grammar. Each is OCRed, validated, segmented into concepts and affirmations, then turned into generative rules.
The pilot source. Full pipeline coverage: OCR validation, semantic tokenisation, concept clustering, generative-rule extraction, formal-grammar synthesis.
74 Hindustani rāgas with structural transcriptions. Extracted via grammar pipeline (Phase 1d-F). Critical edition rendering not yet wired.
Empirical ethnomusicology — microtonal measurements, performance analysis. Extracted (Phase 1d-F). Rendering pending.