Text Generation for Brazilian Portuguese: the Surface Realization Task
نویسندگان
چکیده
Despite the growing interest in NLP focused on the Brazilian Portuguese language in recent years, its obvious counterpart – Natural Language Generation (NLG) – remains in that case a little-explored research field. In this paper we describe preliminary results of a first project of this kind, addressing the issue of surface realization for Brazilian Portuguese. Our approach, which may be particularly suitable to simpler NLG applications in which a domain corpus of the most likely output sentences happens to be available, is in principle adaptable to many closely-related languages, and paves the way to further NLG research focused on Romance languages in general.
منابع مشابه
‘Minor’ Languages, ‘Broken’ Translations: On Brazilian Reworkings of an Albanian Novel
This essay approaches the challenges of global translation in the 21st century from what might still be considered a somewhat uncommon example: a direct translation of Ismail Kadaré's 1978 novel Prill e thyër (Broken April) from the original Albanian into Brazilian Portuguese in 2001. Not only does it examine and compare lexical elements in the source and target texts and the usage of translato...
متن کاملTune or Text? Tune-text accommodation strategies in Portuguese
In Portuguese, different strategies for dealing with tune-text accommodation have been reported. However, no systematic research has been conducted exploring crucial cases of complex nuclear melodies realized in nuclear words with final stress, as in yes-no questions. Based on reading and semispontaneous data from ten regions in Brazil and eleven regions in Portugal, this study reveals that Bra...
متن کاملSupporting the Adaptation of Texts for Poor Literacy Readers: a Text Simplification Editor for Brazilian Portuguese
In this paper we investigate the task of text simplification for Brazilian Portuguese. Our purpose is three-fold: to introduce a simplification tool for such language and its underlying development methodology, to present an on-line authoring system of simplified text based on the previous tool, and finally to discuss the potentialities of such technology for education. The resources and tools ...
متن کاملA model of segment (and pause) duration generation for Brazilian Portuguese text-to-speech synthesis
This work presents and evaluates a model of segmental duration generation for Brazilian Portuguese where the notion of macrorhythmic unit is the starting point to drastically simplify duration assignment and to allow pause insertion as an integrated procedure of generation. This model is preferred to random assignment with the same error distribution. Some aspects of rhythm phonetics and phonol...
متن کاملRDF2PT: Generating Brazilian Portuguese Texts from RDF Data
The generation of natural language from Resource Description Framework (RDF) data has recently gained significant attention due to the continuous growth of Linked Data. A number of these approaches generate natural language in languages other than English, however, no work has been proposed to generate Brazilian Portuguese texts out of RDF. We address this research gap by presenting RDF2PT, an ...
متن کامل