International Workshop on Spoken Language Translation (IWSLT) 2005
Pittsburgh, PA, USA
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.
Full Paper Presentation
Bibliographic reference. Zhang, Ruiqiang / Kikui, Genichiro / Yamamoto, Hirofumi / Lo, Wai-Kit (2005): "A decoding algorithm for word lattice translation in speech translation", In IWSLT-2005, 23-29.