This paper is concerned with speech-to-speech translation that is sensitive to paralinguistic information. From the many different possible paralinguistic features to handle, in this paper we chose duration and power as a first step, proposing a method that can translate these features from input speech to the output speech in continuous space. This is done in a simple and language-independent fashion by training a regression model that maps source language duration and power information into the target language. We evaluate the proposed method on a digit translation task and show that paralinguistic information in input speech appears in output speech, and that this information can be used by target language speakers to detect emphasis.
Cite as: Kano, T., Sakti, S., Takamichi, S., Neubig, G., Toda, T., Nakamura, S. (2012) A method for translation of paralinguistic information. Proc. International Workshop on Spoken Language Translation (IWSLT 2012), 158-163
@inproceedings{kano12_iwslt,
author={Takatomo Kano and Sakriani Sakti and Shinnosuke Takamichi and Graham Neubig and Tomoki Toda and Satoshi Nakamura},
title={{A method for translation of paralinguistic information}},
year=2012,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2012)},
pages={158--163}
}