skip to main content

ComposeInStyle: Music composition with and without Style Transfer

Mukherjee, Sreetama ; Mulimani, Manjunath

Expert systems with applications, 2022-04, Vol.191, p.116195, Article 116195 [Periódico revisado por pares]

New York: Elsevier Ltd

Texto completo disponível

Citações Citado por
  • Título:
    ComposeInStyle: Music composition with and without Style Transfer
  • Autor: Mukherjee, Sreetama ; Mulimani, Manjunath
  • Assuntos: Classification ; Classifiers ; Composers ; Composition ; Datasets ; Evaluation ; Generative Adversarial Networks (GAN) ; Hybrid model ; Music ; Music composer classification ; Musical Instrument Digital Interface (MIDI) ; Piano rolls ; Style transfer
  • É parte de: Expert systems with applications, 2022-04, Vol.191, p.116195, Article 116195
  • Descrição: Every music composition has a composer at the core of its building block, molding it into a style of their own. The creative compositional style of a composer varies dynamically with every composer which is perishable and inimitable but cannot be preserved. This paper proposes a hybrid style transfer model which is an end-to-end approach to transfer style, compose as well as evaluate the style transfer accuracy in the domain of the target music composer. The paper focuses on 3 composer maestros Liszt, Chopin and Schubert taken from the Maestro dataset. The main step of music composer style transfer is accomplished in 3 systematic steps, training composer classifiers according to feature sets, generating a model of a composition in a particular composer’s style from noise, and finally style transfer and its evaluation. The 3 steps are interconnected in a way that each step is the building block for the next step. The GAN architecture of the style transfer step is built out of the GAN architecture of the second step. The compositions generated from each architecture is finally evaluated by the common pre-trained classifier of the first step. The highest accuracy obtained is 80% for composer classification using the maestro dataset, 77.27% for the classification of the generated style transferred version of a composition into the target composer class using the pre-trained classifiers. •Music composer classification with audio features.•Unique composer specific composition generation from noise.•Style transfer from one composer style to another with the proposed model.•Evaluation of generated style transferred compositions with pretrained classifiers.
  • Editor: New York: Elsevier Ltd
  • Idioma: Inglês

Buscando em bases de dados remotas. Favor aguardar.