This paper addresses the issue of modeling the discourse nature of lyrics and presented the first study aiming at capturing the two common discourse-related notions: storylines and themes. We assume that a storyline is a chain of transitions over topics of segments and a song has at least one entire theme. We then hypothesize that transitions over topics of lyric segments can be captured by a probabilistic topic model which incorporates a distribution over transitions of latent topics and that such a distribution of topic transitions is affected by the theme of lyrics. Aiming to test those hypotheses, this study conducts experiments on the word prediction and segment order prediction tasks exploiting a large-scale corpus of popular music lyrics for both English and Japanese (around 100 thousand songs). The findings we gained from these experiments can be summarized into two respects. First, the models with topic transitions significantly outperformed the model without topic transitions in word prediction. This result indicates that typical storylines included in our lyrics datasets were effectively captured as a probabilistic distribution of transitions over latent topics of segments. Second, the model incorporating a latent theme variable on top of topic transitions outperformed the models without such variables in both word prediction and segment order prediction. From this result, we can conclude that considering the notion of theme does contribute to the modeling of storylines of lyrics.
Kento WATANABE
Tohoku University
Yuichiroh MATSUBAYASHI
Tohoku University
Kentaro INUI
Tohoku University,RIKEN Center for Advanced Intelligence Project
Satoru FUKAYAMA
National Institute of Advanced Industrial Science and Technology (AIST)
Tomoyasu NAKANO
National Institute of Advanced Industrial Science and Technology (AIST)
Masataka GOTO
National Institute of Advanced Industrial Science and Technology (AIST)
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Kento WATANABE, Yuichiroh MATSUBAYASHI, Kentaro INUI, Satoru FUKAYAMA, Tomoyasu NAKANO, Masataka GOTO, "Modeling Storylines in Lyrics" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 4, pp. 1167-1179, April 2018, doi: 10.1587/transinf.2017EDP7188.
Abstract: This paper addresses the issue of modeling the discourse nature of lyrics and presented the first study aiming at capturing the two common discourse-related notions: storylines and themes. We assume that a storyline is a chain of transitions over topics of segments and a song has at least one entire theme. We then hypothesize that transitions over topics of lyric segments can be captured by a probabilistic topic model which incorporates a distribution over transitions of latent topics and that such a distribution of topic transitions is affected by the theme of lyrics. Aiming to test those hypotheses, this study conducts experiments on the word prediction and segment order prediction tasks exploiting a large-scale corpus of popular music lyrics for both English and Japanese (around 100 thousand songs). The findings we gained from these experiments can be summarized into two respects. First, the models with topic transitions significantly outperformed the model without topic transitions in word prediction. This result indicates that typical storylines included in our lyrics datasets were effectively captured as a probabilistic distribution of transitions over latent topics of segments. Second, the model incorporating a latent theme variable on top of topic transitions outperformed the models without such variables in both word prediction and segment order prediction. From this result, we can conclude that considering the notion of theme does contribute to the modeling of storylines of lyrics.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017EDP7188/_p
Copy
@ARTICLE{e101-d_4_1167,
author={Kento WATANABE, Yuichiroh MATSUBAYASHI, Kentaro INUI, Satoru FUKAYAMA, Tomoyasu NAKANO, Masataka GOTO, },
journal={IEICE TRANSACTIONS on Information},
title={Modeling Storylines in Lyrics},
year={2018},
volume={E101-D},
number={4},
pages={1167-1179},
abstract={This paper addresses the issue of modeling the discourse nature of lyrics and presented the first study aiming at capturing the two common discourse-related notions: storylines and themes. We assume that a storyline is a chain of transitions over topics of segments and a song has at least one entire theme. We then hypothesize that transitions over topics of lyric segments can be captured by a probabilistic topic model which incorporates a distribution over transitions of latent topics and that such a distribution of topic transitions is affected by the theme of lyrics. Aiming to test those hypotheses, this study conducts experiments on the word prediction and segment order prediction tasks exploiting a large-scale corpus of popular music lyrics for both English and Japanese (around 100 thousand songs). The findings we gained from these experiments can be summarized into two respects. First, the models with topic transitions significantly outperformed the model without topic transitions in word prediction. This result indicates that typical storylines included in our lyrics datasets were effectively captured as a probabilistic distribution of transitions over latent topics of segments. Second, the model incorporating a latent theme variable on top of topic transitions outperformed the models without such variables in both word prediction and segment order prediction. From this result, we can conclude that considering the notion of theme does contribute to the modeling of storylines of lyrics.},
keywords={},
doi={10.1587/transinf.2017EDP7188},
ISSN={1745-1361},
month={April},}
Copy
TY - JOUR
TI - Modeling Storylines in Lyrics
T2 - IEICE TRANSACTIONS on Information
SP - 1167
EP - 1179
AU - Kento WATANABE
AU - Yuichiroh MATSUBAYASHI
AU - Kentaro INUI
AU - Satoru FUKAYAMA
AU - Tomoyasu NAKANO
AU - Masataka GOTO
PY - 2018
DO - 10.1587/transinf.2017EDP7188
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2018
AB - This paper addresses the issue of modeling the discourse nature of lyrics and presented the first study aiming at capturing the two common discourse-related notions: storylines and themes. We assume that a storyline is a chain of transitions over topics of segments and a song has at least one entire theme. We then hypothesize that transitions over topics of lyric segments can be captured by a probabilistic topic model which incorporates a distribution over transitions of latent topics and that such a distribution of topic transitions is affected by the theme of lyrics. Aiming to test those hypotheses, this study conducts experiments on the word prediction and segment order prediction tasks exploiting a large-scale corpus of popular music lyrics for both English and Japanese (around 100 thousand songs). The findings we gained from these experiments can be summarized into two respects. First, the models with topic transitions significantly outperformed the model without topic transitions in word prediction. This result indicates that typical storylines included in our lyrics datasets were effectively captured as a probabilistic distribution of transitions over latent topics of segments. Second, the model incorporating a latent theme variable on top of topic transitions outperformed the models without such variables in both word prediction and segment order prediction. From this result, we can conclude that considering the notion of theme does contribute to the modeling of storylines of lyrics.
ER -