Yamato Takahashi

Learn More
This paper introduces a new examplebased method of machine translation in which the examples need not be direct translations. The system will weed out strange examples during translation, allowing the use of currently available sentence aligned corpora as data. Rulebased modules are used where appropriate. A prototype Japanese-to-English system has been(More)
This paper describes the continuation of work on tagging text used in research into Japanese-to-English machine translation, first described in (Bond et al. 1995). As part of ongoing research into Japanese-to-English machine translation we have, for some time, been collecting various aligned parallel texts. The majority of these have been Japanese original(More)
This paper discusses the stimulation of the subject's strong emotion in listening music and the relationships between the impression or the strong emotion and acoustic feature values. Impressions of music are classified into five types: "charity," "nobility," "nostalgia," "passion," and "joy." First, using fifteen classical music pieces, an impression(More)
This paper shows the system mutually retrieving pictures and sound clips based on their impressions. Pictures (sound clips, respectively) are retrieved by specifying a music clip (picture). Pictures (sound clips) having similar impression to a sound clip (picture) are retrieved. No words are required in retrieving them. To this end, the scores of the(More)
  • 1