Régis Quélavoine

We don’t have enough information about this author to calculate their statistics. If you think this is an error let us know.
Learn More
In this paper, aspeech rescoring system is developed on a set of phonetic h ypotheses produced by a bottom-up knowledge-based decoder. An original m ethod to automatically c o m pute a fuzzy membership function from top-down acoustic rules statistics is compared with a possibilistic measure. To aggregate the fuzzy degrees into a phonetic s c o re, a(More)
  • 1