Neural Poetry Translation

@inproceedings{Ghazvininejad2018NeuralPT,
  title={Neural Poetry Translation},
  author={Marjan Ghazvininejad and Yejin Choi and Kevin Knight},
  booktitle={NAACL},
  year={2018}
}
We present the first neural poetry translation system. Unlike previous works that often fail to produce any translation for fixed rhyme and rhythm patterns, our system always translates a source text to an English poem. Human evaluation of the translations ranks the quality as acceptable 78.2% of the time. 

Figures, Tables, and Topics from this paper

First Steps Towards the Creation of a Poetry Translation Mapping System
The translation of poetry is a complex, multifaceted challenge: the translated text should communicate the same meaning, similar metaphoric expressions, and also match the style and prosody of theExpand
The Challenges of Using Neural Machine Translation for Literature
In this paper, we adapt state-of-the-art neural machine translation (NMT) systems to literary content and use them to translate fiction stories from English to Russian and from German to English. WeExpand
Automatic Machine Translation of Poetry and a Low-Resource Language Pair
  • I. Dunder, S. Seljan, M. Pavlovski
  • Computer Science
  • 2020 43rd International Convention on Information, Communication and Electronic Technology (MIPRO)
  • 2020
TLDR
The aim of this paper is to examine the usability of machine translation for poetry and a low-resource language pair, such as Croatian-German, and demonstrates the effectiveness of machinetranslation of poetry with regard to special automatic quality metrics. Expand
Rhetorically Controlled Encoder-Decoder for Modern Chinese Poetry Generation
TLDR
This paper proposes a rhetorically controlled encoder-decoder for modern Chinese poetry generation that relies on a continuous latent variable as a rhetoric controller to capture various rhetorical patterns in an encoder, and then incorporates rhetoric-based mixtures while generatingmodern Chinese poetry. Expand
Learning Rhyming Constraints using Structured Adversaries
TLDR
This work proposes an alternate approach that uses a structured discriminator to learn a poetry generator that directly captures rhyming constraints in a generative adversarial setup and is able to successfully learn rhyming patterns in two different English poetry datasets. Expand
Automatic Dialect Adaptation in Finnish and its Effect on Perceived Creativity
TLDR
The results suggest that the more the dialect deviates from the standard Finnish, the lower scores people tend to give on an existing evaluation metric. Expand
Syllabification of the Divine Comedy
TLDR
An online vocabulary is provided, for each word, information about its syllabification, the location of the tonic accent, and the aforementioned synalephe propensity, on the left and right sides, which opens a wide range of exciting perspectives, from the possibility of automatic learning syllabified words and verses to the improvement of generative models. Expand
Neural Language Generation: Formulation, Methods, and Evaluation
TLDR
There is no standard way to assess the quality of text produced by these generative models, which constitutes a serious bottleneck towards the progress of the field, so this survey will provide an informative overview of formulations, methods, and assessments of neural natural language generation. Expand
Les lettres et la machine : un état de l’art en traduction littéraire automatique (Machines in the humanities: current state of the art in literary machine translation)
Étant donné la récente vague d’intérêt pour la traduction littéraire automatique, cet article vise à recenser les travaux déjà parus sur le sujet, tout en partageant quelques prises de position surExpand
A General-Purpose Algorithm for Constrained Sequential Inference
TLDR
This work presents a principled approach for incorporating constraints into sequential inference algorithms, and expresses constraints using an automaton, which is traversed in lock-step during inference, guiding the search to valid outputs. Expand
...
1
2
...

References

SHOWING 1-10 OF 16 REFERENCES
“Poetic” Statistical Machine Translation: Rhyme and Meter
TLDR
The ability to produce translations with meter and rhyme for phrase-based MT is implemented, and whether the hypothesis space of such a system is flexible enough to accomodate such constraints is examined. Expand
Automatic Analysis of Rhythmic Poetry with Applications to Generation and Translation
TLDR
This work applies unsupervised learning to reveal word-stress patterns in a corpus of raw poetry to generate English love poetry, and translates Italian poetry into English, choosing target realizations that conform to desired rhythmic patterns. Expand
Generating Topical Poetry
TLDR
Hafez is a program that generates any number of distinct poems on a usersupplied topic that obeys rhythmic and rhyme constraints and shows its generality with respect to language and poetic form. Expand
PoeTryMe : a versatile platform for poetry generation
TLDR
A prototype was implemented based on PoeTryMe to generate Portuguese poetry, using natural language processing resources for this language, and patterns that denote semantic relations in human-created poetry. Expand
Automatically Generating Rhythmic Verse with Neural Networks
TLDR
Two novel methodologies for the automatic generation of rhythmic poetry in a variety of forms are proposed that demonstrate that participants consider machine-generated poems to be written by humans 54% of the time and that participants rated a machine- generated poem to be the best amongst all evaluated. Expand
Generating Chinese Classical Poems with Statistical Machine Translation Models
TLDR
A statistical approach to generation of Chinese classical poetry and a novel method to automatically evaluate poems are described and the effectiveness of the BLEU metric for evaluation is demonstrated with a Novel method of generating diverse references. Expand
Generating Chinese Classical Poems with RNN Encoder-Decoder
TLDR
A novel system based on the RNN Encoder-Decoder structure to generate quatrains (Jueju in Chinese), with a topic word as input that can jointly learn semantic meaning within a single line, semantic relevance among lines in a poem, and the use of structural, rhythmical and tonal patterns, without utilizing any constraint templates. Expand
i, Poet: Automatic Chinese Poetry Composition through a Generative Summarization Framework under Constrained Optimization
TLDR
This paper forms the poetry composition task as an optimization problem based on a generative summarization framework under several constraints, and retrieves candidate terms out of a large poem corpus, and orders these terms to fit into poetry formats, satisfying tonal and rhythm requirements. Expand
Chinese Poetry Generation with Recurrent Neural Networks
TLDR
A model for Chinese poem generation based on recurrent neural networks which is ideally suited to capturing poetic content and form is proposed which outperforms competitive Chinese poetry generation systems using both automatic and manual evaluation methods. Expand
A Survey on Intelligent Poetry Generation: Languages, Features, Techniques, Reutilisation and Evaluation
TLDR
This paper surveys intelligent poetry generators around a set of relevant axis for poetry generation – targeted languages, form and content features, techniques, reutilisation of material, and evaluation – and aims to organise work developed on this topic so far. Expand
...
1
2
...