Modeling Baroque Two-Part Counterpoint ...
Type de document :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Titre :
Modeling Baroque Two-Part Counterpoint with Neural Machine Translation
Auteur(s) :
Nichols, Eric [Auteur]
Kalonaris, Stefano [Auteur]
RIKEN Center for Advanced Intelligence Project [Tokyo] [RIKEN AIP]
Micchi, Gianluca [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Algomus
Aljanaki, Anna [Auteur]
University of Tartu
Kalonaris, Stefano [Auteur]
RIKEN Center for Advanced Intelligence Project [Tokyo] [RIKEN AIP]
Micchi, Gianluca [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Algomus
Aljanaki, Anna [Auteur]
University of Tartu
Titre de la manifestation scientifique :
International Computer Music Conference (ICMC 2020)
Ville :
Santiago
Pays :
Chili
Date de début de la manifestation scientifique :
2020
Discipline(s) HAL :
Sciences de l'Homme et Société/Musique, musicologie et arts de la scène
Informatique [cs]/Son [cs.SD]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Son [cs.SD]
Informatique [cs]/Apprentissage [cs.LG]
Résumé en anglais : [en]
We propose a system for contrapuntal music generation based on a Neural Machine Translation (NMT) paradigm. We consider Baroque counterpoint and are interested in modeling the interaction between any two given parts as a ...
Lire la suite >We propose a system for contrapuntal music generation based on a Neural Machine Translation (NMT) paradigm. We consider Baroque counterpoint and are interested in modeling the interaction between any two given parts as a mapping between a given source material and an appropriate target material. Like in translation, the former imposes some constraints on the latter, but doesn’t define it completely. We collate and edit a bespoke dataset of Baroque pieces, use it to train an attention-based neural network model, and evaluate the generated output via BLEU score and musicological analysis. We show that our model is able to respond with some idiomatic trademarks, such as imitation and appropriate rhythmic offset, although it falls short of having learned stylistically correct contrapuntal motion (e.g., avoidance of parallel fifths) or stricter imitative rules, such as canonLire moins >
Lire la suite >We propose a system for contrapuntal music generation based on a Neural Machine Translation (NMT) paradigm. We consider Baroque counterpoint and are interested in modeling the interaction between any two given parts as a mapping between a given source material and an appropriate target material. Like in translation, the former imposes some constraints on the latter, but doesn’t define it completely. We collate and edit a bespoke dataset of Baroque pieces, use it to train an attention-based neural network model, and evaluate the generated output via BLEU score and musicological analysis. We show that our model is able to respond with some idiomatic trademarks, such as imitation and appropriate rhythmic offset, although it falls short of having learned stylistically correct contrapuntal motion (e.g., avoidance of parallel fifths) or stricter imitative rules, such as canonLire moins >
Langue :
Anglais
Comité de lecture :
Oui
Audience :
Internationale
Vulgarisation :
Non
Collections :
Source :