Efficient Phrase-table Representation for Machine Translation with Applications to Online MT and Speech Translation

In phrase-based statistical machine translation, the phrase-table requires a large amount of memory. We will present an efficient representation with two key properties: on-demand loading and a prefix tree structure for the source phrases. We will show that this representation scales well to large data tasks and that we are able to store hundreds of millions of phrase pairs in the phrase-table. For the large Chinese–English NIST task, the memory requirements of the phrase-table are reduced to less than 20 MB using the new representation with no loss in translation quality and speed. Additionally, the new representation is not limited to a specific test set, which is important for online or real-time machine translation. One problem in speech translation is the matching of phrases in the input word graph and the phrase-table. We will describe a novel algorithm that effectively solves this combinatorial problem exploiting the prefix tree data structure of the phrase-table. This algorithm enables the use of significantly larger input word graphs in a more efficient way resulting in improved translation quality.