File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Coarse-To-Fine Framework For Music Generation via Generative Adversarial Networks

TitleCoarse-To-Fine Framework For Music Generation via Generative Adversarial Networks
Authors
Issue Date2020
PublisherAssociation for Computing Machinery.
Citation
Proceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial Intelligence (HPCCT & BDAI 2020), Qingdao, China, 3-6 July 2020, p. 192-198 How to Cite?
AbstractAutomatic music generation is highly related to Natural Language Processing (NLP). A current note in melody always depends on its context, just like a word in NLP. Yet the difference is that music is built upon a set of special chords that formulates the skeleton of the melody. To enhance automatic music generation, we propose a two-step adversarial procedure: Step 1 learns to generate chords via a chord generative adversarial networks (GANs); and step 2 trains a melody GAN to generate music for which the input is conditioned on the chords produced through the first step. Under such a two-step procedure, the chords generated in the first step formulate a basic framework of the music, which can theoretically and practically improve the performance of melody generation in the second step. Experiments demonstrate that such a cascading process is able to generate high-quality music samples with both acoustical and music theoretical guarantees.
Persistent Identifierhttp://hdl.handle.net/10722/294829
ISBN

 

DC FieldValueLanguage
dc.contributor.authorMa, D-
dc.contributor.authorBin, L-
dc.contributor.authorQiao, X-
dc.contributor.authorCao, D-
dc.contributor.authorYin, G-
dc.date.accessioned2020-12-21T11:49:09Z-
dc.date.available2020-12-21T11:49:09Z-
dc.date.issued2020-
dc.identifier.citationProceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial Intelligence (HPCCT & BDAI 2020), Qingdao, China, 3-6 July 2020, p. 192-198-
dc.identifier.isbn9781450375603-
dc.identifier.urihttp://hdl.handle.net/10722/294829-
dc.description.abstractAutomatic music generation is highly related to Natural Language Processing (NLP). A current note in melody always depends on its context, just like a word in NLP. Yet the difference is that music is built upon a set of special chords that formulates the skeleton of the melody. To enhance automatic music generation, we propose a two-step adversarial procedure: Step 1 learns to generate chords via a chord generative adversarial networks (GANs); and step 2 trains a melody GAN to generate music for which the input is conditioned on the chords produced through the first step. Under such a two-step procedure, the chords generated in the first step formulate a basic framework of the music, which can theoretically and practically improve the performance of melody generation in the second step. Experiments demonstrate that such a cascading process is able to generate high-quality music samples with both acoustical and music theoretical guarantees.-
dc.languageeng-
dc.publisherAssociation for Computing Machinery.-
dc.relation.ispartofProceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial Intelligence (HPCCT & BDAI 2020)-
dc.rightsProceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial Intelligence (HPCCT & BDAI 2020). Copyright © Association for Computing Machinery.-
dc.titleCoarse-To-Fine Framework For Music Generation via Generative Adversarial Networks-
dc.typeConference_Paper-
dc.identifier.emailYin, G: gyin@hku.hk-
dc.identifier.authorityYin, G=rp00831-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1145/3409501.3409534-
dc.identifier.hkuros320601-
dc.identifier.spage192-
dc.identifier.epage198-
dc.publisher.placeNew York, NY-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats