2015/09/16 by Francisco Gómez, Joaquín Mora, Gómez, Francisco +6
Arts and Humanities · Computer Science · #Diverse Musicological Studies #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Music and Audio Processing #Musicology and Musical Analysis #Sound (cs.SD) #Theater, Performance, and Music History
paper · pdf · doi:10.48550/arxiv.1509.04956
openalex publication_date 2015/09/16 · openalex created_date 2022/09/27 · openalex updated_date 2026/07/28
This work focuses on the topic of melodic characterization and similarity in\na specific musical repertoire: a cappella flamenco singing, more specifically\nin debla and martinete styles. We propose the combination of manual and\nautomatic description. First, we use a state-of-the-art automatic transcription\nmethod to account for general melodic similarity from music recordings. Second,\nwe define a specific set of representative mid-level melodic features, which\nare manually labeled by flamenco experts. Both approaches are then contrasted\nand combined into a global similarity measure. This similarity measure is\nassessed by inspecting the clusters obtained through phylogenetic algorithms\nalgorithms and by relating similarity to categorization in terms of style.\nFinally, we discuss the advantage of combining automatic and expert annotations\nas well as the need to include repertoire-specific descriptions for meaningful\nmelodic characterization in traditional music collections.\n