Today, word sense disambiguation (WSD) is an important technique for many natural language processing (NLP) applications such as grammatical analysis, content analysis, information retrieval and machine translation. Among them, the WSD technique is used for machine translation to find the correct sense of a word in a specific context. In machine translation, the input sentences in the source la...

In the spoken language translation pipeline, machine translation systems that are trained solely on written bitexts are often unable to recover from speech recognition errors due to the mismatch in training data. We propose a novel technique to simulate the errors generated by an ASR system, using the ASR system’s pronunciation dictionary and language model. Lexical entries in the pronunciation...

Towards smooth and natural multilingual speech communication, it is desired to develop simultaneous interpretation systems. This paper proposes a technique for incremental Japanese generation in EnglishJapanese simultaneous machine interpretation. Generally, the difference of word-order in both languages causes a fatal problem in incremental translation between English and Japanese. In our appr...

We give a translation of the visual description technique DiCharts to the synchronous language PURR. As a result, the design of a system can start with a graphical description technique at a very high-level and can then be automatically translated to the synchronous language PURR. This translation allows the further synthesis of the system and in particular, its verification.

The paper presents a novel technique for speech translation using hierarchical phrasedbased statistical machine translation (HPBSMT). The system is based on translation of speech from phone sequences as opposed to conventional approach of speech translation from word sequences. The technique facilitates speech translation by allowing a machine translation (MT) system to access to phonetic infor...

We propose and evaluate a simple technique of “reverse self-training” for statistical machine translation. The technique allows to extend target-side vocabulary of the MT system using target-side monolingual data and it is especially aimed at translation to morphologically rich languages.