| 摘要 |
PURPOSE:To provide a sound recognition device capable of reducing a processing amount more greatly compared with a conventional example and improving a speed for processing sound recognition. CONSTITUTION:In a sound recognition device provided with sound recognition parts 4 and 5 for recognizing the sound of an uttered voice sentence made of an input character string, a non sound area detecting part 30 for outputting a detecting signal by detecting, based on an input uttered voice sentence, at least one of borders between a pause and a redundant word and a phrase or a passage is provided, and the sound recognition parts 4 and 5 executes, when the detecting signal is inputted, sound recognition processing by compressing a plurality of sound recognition candidates syntactically functioning in the same way to one sound recognition candidate. |
