摘要 |
PROBLEM TO BE SOLVED: To provide a voice recognition system capable of accurately changing a threshold for detecting a speech section of a user, and to provide a voice recognizer. SOLUTION: The voice recognition system 1 includes a television set 21, a voice input part 11a, and a speech section detection part 13. The speech section detection part 13 detects the speech section of the user from voice input from the voice input part 11a. Furthermore, a speech section detection part 14 has the threshold for detecting the speech section of the user from the voice input with the voice input part 11a, detects a voice part having sound pressure exceeding the threshold as the speech section of the user, inputs a voice signal from the television set 21, and changes the threshold in response to the input voice signal. COPYRIGHT: (C)2009,JPO&INPIT
|