VirtuosoNet: A Hierarchical RNN-based System for Modeling Expressive Piano Performance

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 221
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorJeong, Dasaemko
dc.contributor.authorKWON, TAEGYUNko
dc.contributor.authorKim, Yoojinko
dc.contributor.authorNam, Juhanko
dc.date.accessioned2019-12-18T03:20:43Z-
dc.date.available2019-12-18T03:20:43Z-
dc.date.created2019-06-12-
dc.date.created2019-06-12-
dc.date.created2019-06-12-
dc.date.issued2019-11-04-
dc.identifier.citationThe 20th International Society for Music Information Retrieval Conference (ISMIR), pp.908 - 915-
dc.identifier.urihttp://hdl.handle.net/10203/269876-
dc.description.abstractIn this paper, we present our application of deep neural network to modeling piano performance, which imitates the expressive control of tempo, dynamics, articulations and pedaling from pianists. Our model consists of recurrent neural networks with hierarchical attention and conditional variational autoencoder. The model takes a sequence of note-level score features extracted from MusicXML as input and predicts piano performance features of the corresponding notes. To render musical expressions consistently over long-term sections, we first predict tempo and dynamics in measure-level and, based on the result, refine them in note-level. The evaluation through listening test shows that our model achieves a more human-like expressiveness compared to previous models.We also share the dataset we used for the experiment.-
dc.languageEnglish-
dc.publisherInternational Society for Music Information Retrieval Conference (ISMIR)-
dc.titleVirtuosoNet: A Hierarchical RNN-based System for Modeling Expressive Piano Performance-
dc.typeConference-
dc.identifier.scopusid2-s2.0-85087094425-
dc.type.rimsCONF-
dc.citation.beginningpage908-
dc.citation.endingpage915-
dc.citation.publicationnameThe 20th International Society for Music Information Retrieval Conference (ISMIR)-
dc.identifier.conferencecountryNE-
dc.identifier.conferencelocationDelft-
dc.contributor.localauthorNam, Juhan-
Appears in Collection
GCT-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0