TY - GEN
T1 - Modeling structural topic transitions for automatic lyrics generation
AU - Watanabe, Kento
AU - Matsubayashi, Yuichiroh
AU - Inui, Kentaro
AU - Goto, Masataka
N1 - Publisher Copyright:
Copyright 2014 by Kento Watanabe, Yuichiroh Matsubayashi, Kentaro Inui, and Masataka Goto.
PY - 2014
Y1 - 2014
N2 - By adopting recent advances in music creation technologies, such as digital audio workstations and singing voice synthesizers, people can now create songs in their personal computers. Computers can also assist in creating lyrics or generating them automatically, although this aspect has been less thoroughly researched and is limited to rhyme and meter. This study focuses on the structural relations in Japanese lyrics. We present novel generation models that capture the topic transitions between units peculiar to the lyrics, such as verse/chorus and line. These transitions are modeled by a Hidden Markov Model (HMM) for representing topics and topic transitions. To verify that our models generate contextsuitable lyrics, we evaluate the models using a log probability of lyrics generation and fill-in-the-blanks-type test. The results show that the language model is far more effective than HMM-based models, but the HMM-based approach successfully captures the inter-verse/chorus and inter-line relations. In the result of experimental evaluation, our approach captures the inter-verse/chorus and inter-line relations.
AB - By adopting recent advances in music creation technologies, such as digital audio workstations and singing voice synthesizers, people can now create songs in their personal computers. Computers can also assist in creating lyrics or generating them automatically, although this aspect has been less thoroughly researched and is limited to rhyme and meter. This study focuses on the structural relations in Japanese lyrics. We present novel generation models that capture the topic transitions between units peculiar to the lyrics, such as verse/chorus and line. These transitions are modeled by a Hidden Markov Model (HMM) for representing topics and topic transitions. To verify that our models generate contextsuitable lyrics, we evaluate the models using a log probability of lyrics generation and fill-in-the-blanks-type test. The results show that the language model is far more effective than HMM-based models, but the HMM-based approach successfully captures the inter-verse/chorus and inter-line relations. In the result of experimental evaluation, our approach captures the inter-verse/chorus and inter-line relations.
UR - http://www.scopus.com/inward/record.url?scp=84994113253&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84994113253&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84994113253
T3 - Proceedings of the 28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014
SP - 422
EP - 431
BT - Proceedings of the 28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014
A2 - Boonkwan, Prachya
A2 - Aroonmanakun, Wirote
A2 - Supnithi, Thepchai
PB - Faculty of Pharmaceutical Sciences, Chulalongkorn University
T2 - 28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014
Y2 - 12 December 2014 through 14 December 2014
ER -