DC Field | Value | Language |
---|---|---|
dc.contributor.author | Doh, Seungheon | ko |
dc.contributor.author | Lee, Junwon | ko |
dc.contributor.author | Nam, Juhan | ko |
dc.date.accessioned | 2021-12-15T06:47:55Z | - |
dc.date.available | 2021-12-15T06:47:55Z | - |
dc.date.created | 2021-12-10 | - |
dc.date.issued | 2021-11-12 | - |
dc.identifier.citation | 2nd Workshop on NLP for Music and Spoken Audio(NLP4MuSA 2021) | - |
dc.identifier.uri | http://hdl.handle.net/10203/290674 | - |
dc.description.abstract | We propose a machine-translation approach to automatically generate a playlist title from a set of music tracks. We take a sequence of track IDs as input and a sequence of words in a playlist title as output, adapting the sequence-to-sequence framework based on Recurrent Neural Network (RNN) and Transformer to the music data. Considering the orderless na-ture of music tracks in a playlist, we propose two techniques that remove the order of the input sequence. One is data augmentation by shuffling and the other is deleting the posi-tional encoding. We also reorganize the exist-ing music playlist datasets to generate phrase-level playlist titles. The result shows that the Transformer models generally outperform the RNN model. Also, removing the order of input sequence improves the performance further. | - |
dc.language | English | - |
dc.publisher | NLP for Music and Spoken Audio | - |
dc.title | Music Playlist Title Generation: A Machine-Translation Approach | - |
dc.type | Conference | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | 2nd Workshop on NLP for Music and Spoken Audio(NLP4MuSA 2021) | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Virtual | - |
dc.contributor.localauthor | Nam, Juhan | - |
dc.contributor.nonIdAuthor | Lee, Junwon | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.