Automatic Analysis of Rhythmic Poetry with Applications to Generation and Translation

We employ statistical methods to analyze, generate, and translate rhythmic poetry. We first apply unsupervised learning to reveal word-stress patterns in a corpus of raw poetry. We then use these word-stress patterns, in addition to rhyme and discourse models, to generate English love poetry. Finally, we translate Italian poetry into English, choosing target realizations that conform to desired rhythmic patterns.

• Type: Journal
• Year: 2010
• Where: EMNLP
• Authors: Erica Greene, Tugba Bodrumlu, Kevin Knight
• Keywords: EMNLP 2010, English Love Poetry, Natural Language Processing, Poetry, Word-stress Patterns