We propose a novel statistical machine translation decoding algorithm for speech translation to improve speech translation quality. The algorithm can translate the speech recognition word lattice, where more hypotheses are utilized to bypass the misrecognized single-best hypotheses. We also show that a speech recognition confidence measure, implemented by posterior probability, is effective to improve speech translation. The proposed techniques were tested in a Japanese-to-English speech translation task. The experimental results demonstrate the improved speech translation performance by the proposed techniques.
Cite as: Zhang, R., Kikui, G., Yamamoto, H., Lo, W.-K. (2005) A decoding algorithm for word lattice translation in speech translation. Proc. International Workshop on Spoken Language Translation (IWSLT 2005), 23-29
@inproceedings{zhang05_iwslt,
author={Ruiqiang Zhang and Genichiro Kikui and Hirofumi Yamamoto and Wai-Kit Lo},
title={{A decoding algorithm for word lattice translation in speech translation}},
year=2005,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2005)},
pages={23--29}
}