MISTRAL: a Statistical Machine Translation Decoder for Speech Recognition Lattices

This paper presents MISTRAL, an open source statistical machine translation decoder dedicated to spoken language translation. While typical machine translation systems take a written text as input, MISTRAL translates word lattices produced by automatic speech recognition systems. The lattices are translated in two passes using a phrase-based model. Our experiments reveal an improvement in BLEU when translating lattices instead of sentences returned by a speech recognition system.

• Type: Conference
• Year: 2008
• Where: LREC
• Authors: Alexandre Patry, Philippe Langlais
• Keywords: Education, LREC 2008, Machine Translation, Machine Translation Decoder, Speech Recognition System