13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Unsupervised Deep Belief Features for Speech Translation

Sameer Maskey, Bowen Zhou

IBM T.J. Watson Research Center, Yorktown Heights, NY, USA

We present a novel formalism for introducing deep belief features to Hierarchical Machine Translation Model. The deep features are generated by unsupervised training of a deep belief network built with stacked sets of Restricted Boltzmann Machines. We show that our new deep feature based hierarchical model is significantly better than the baseline hierarchical model with gains for two different languages pairs in two different data size settings. We obtain absolute BLEU score improvement of +1.13 on Dari-to-English and +0.66 on English-to-Dari Transtac Evaluation task. We also observe gains on English-to-Chinese translation task.

Full Paper

Bibliographic reference.  Maskey, Sameer / Zhou, Bowen (2012): "Unsupervised deep belief features for speech translation", In INTERSPEECH-2012, 2358-2361.