| 摘要 |
PROBLEM TO BE SOLVED: To improve the recognition rate and recognition precision of person's speaking voice in every kinds of noise environments. SOLUTION: In the case a present noise pattern is determined to be a new noise pattern which is not included in preliminarily set various noise patterns, a speaking voice is collected again by a non-contact type microphone and a contact type microphone, and the voice collected by the contact microphone is recognized, and the voice collected by the non-contact microphone is labeled on the basis of the recognition result, and a sound model for non-contact microphone, which corresponds to the new noise pattern, is generated and stored. COPYRIGHT: (C)2007,JPO&INPIT
|
