Automatic analysis of poetic rhythm is a challenging task that involveslinguistics, literature, and computer science. When the language to be analyzedis known, rule-based systems or data-driven methods can be used. In this paper,we analyze poetic rhythm in English and Spanish. We show that therepresentations of data learned from character-based neural models are moreinformative than the ones from hand-crafted features, and that aBi-LSTM+CRF-model produces state-of-the art accuracy on scansion of poetry intwo languages. Results also show that the information about whole wordstructure, and not just independent syllables, is highly informative forperforming scansion.
展开▼
