Automatic interpretation or automatic translation of speech: concepts, definitions and software architecture

Authors

  • Flávio Freitas Federal University of Uberlândia (UFU), Uberlândia, Brazil.
  • Marileide Esqueda Federal University of Uberlândia (UFU), Uberlândia, Brazil.

DOI:

https://doi.org/10.11606/issn.2317-9511.v29i0p104-145

Keywords:

Machine Interpreting, Speech-to-Speech Machine Translation, Speech Translation Technology

Abstract

Machine Interpreting (MI) or Speech-to-Speech Translation (S2S) is a new technology that converts spoken utterances from one language into another through three functionalities grouped in only one software: Automatic Speech Recognition (ASR), Machine Translation (MT) and Speech Synthesis (or TTS - Text-To-Speech). Firstly presented in 1983, during ITU Telecom Convention, in Geneva, MI consists in a software that enables natural and efficient communication between people from different languages (Lee, 2015). Nevertheless, MI and its components are scarcely studied, especially in Interpreting Studies, generally investigated by the areas of Computer Sciences (Pöchhacker 2004). Following this trend, this paper aims to present the result of a bibliographical research whose proposal was to investigate how MI is conceived by scholars.

Downloads

Download data is not yet available.

Author Biographies

  • Flávio Freitas, Federal University of Uberlândia (UFU), Uberlândia, Brazil.
    Graduate student at Federal University of Uberlândia (UFU)
  • Marileide Esqueda, Federal University of Uberlândia (UFU), Uberlândia, Brazil.
    Professor at Federal University of Uberlândia (UFU)

Published

2017-07-13

Issue

Section

Articles

How to Cite

Freitas, F., & Esqueda, M. (2017). Automatic interpretation or automatic translation of speech: concepts, definitions and software architecture. TradTerm, 29, 104-145. https://doi.org/10.11606/issn.2317-9511.v29i0p104-145