Lexicon C19 Export Package v3.2.1: A Multilingual Dataset of Nineteenth-Century Music Performance Lexicography

Version 3.2.1 of Lexicon C19 is a maintenance release of the v3.2 dataset. It preserves 3,760 rule-selected records and the complete pool of 25,186 parsed candidates from eight nineteenth-century music, dance, biographical, and general-reference traditions, and adds the completed 200-record label-blinded audit workbook, an explicit MIT code-licence file for scripts and executable notebooks, and refreshed publication-analysis outputs. The selected CSV and complete candidate-pool records are unchanged from v3.2. The corpus is machine-generated and structurally validated with limited single-annotator audit evidence; it does not claim segmentation recall, relevance recall, inter-annotator agreement, or gold-standard labels. This work acknowledges support from ANR MELODY (ANR-24-IAS1-0001).