摘要 |
PROBLEM TO BE SOLVED: To provide a speech recognition system that can inform a user of whether a speech analyzing device can understand a speech spoken to a microphone through a user interface part. SOLUTION: A sound pressure/frequency characteristic measuring instrument of the speech recognition part inputs environmental noise in a time zone wherein no speech is inputted from the microphone and measures its sound pressure and a temporal variation quantity of a frequency characteristic. A speech recognition possibility decision device decides that speech recognition is "excellent", "possible", or "impossible". A state informing device informs the user of the decision result of the speech recognition possibility decision device. COPYRIGHT: (C)2004,JPO&NCIPI
|