Publicação

Motiv: a dataset of latent space representations of musical phrase motions

Ver documento

Detalhes bibliográficos
Resumo:This paper introduces Motiv, a dataset of expert saxophonist record- ings illustrating parallel, similar, oblique, and contrary motions. These motions are variations of three phrases from Jesús Villa- Rojo’s “Lamento,” with controlled similarities. The dataset includes 116 audio samples recorded by four tenor saxophonists, each anno- tated with descriptions of motions, musical scores, and latent space vectors generated using the VocalSet RAVE model. Motiv enables the analysis of motion types and their geometric relationships in latent spaces. Our preliminary dataset analysis shows that parallel motions align closely with original phrases, while contrary motions exhibit the largest deviations, and oblique motions show mixed pat- terns. The dataset also highlights the impact of individual performer nuances. Motiv supports a variety of music information retrieval (MIR) tasks, including gesture-based recognition, performance anal- ysis, and motion-driven retrieval. It also provides insights into the relationship between human motion and music, contributing to real-time music interaction and automated performance systems.
Autores principais:Carvalho, Nádia
Outros Autores:Sousa, Jorge; Bernardes, Gilberto; Portovedo, Henrique
Assunto:Music performance Latent space navigation Timbre Musical motions Co-improvisation systems Electroacoustic tape AI
Ano:2025
País:Portugal
Tipo de documento:artigo
Tipo de acesso:acesso restrito
Instituição associada:Universidade de Aveiro
Idioma:inglês
Origem:RIA - Repositório Institucional da Universidade de Aveiro
Descrição
Resumo:This paper introduces Motiv, a dataset of expert saxophonist record- ings illustrating parallel, similar, oblique, and contrary motions. These motions are variations of three phrases from Jesús Villa- Rojo’s “Lamento,” with controlled similarities. The dataset includes 116 audio samples recorded by four tenor saxophonists, each anno- tated with descriptions of motions, musical scores, and latent space vectors generated using the VocalSet RAVE model. Motiv enables the analysis of motion types and their geometric relationships in latent spaces. Our preliminary dataset analysis shows that parallel motions align closely with original phrases, while contrary motions exhibit the largest deviations, and oblique motions show mixed pat- terns. The dataset also highlights the impact of individual performer nuances. Motiv supports a variety of music information retrieval (MIR) tasks, including gesture-based recognition, performance anal- ysis, and motion-driven retrieval. It also provides insights into the relationship between human motion and music, contributing to real-time music interaction and automated performance systems.