摘要 |
PROBLEM TO BE SOLVED: To provide a speech recognition device capable of improving a speech recognition performance. SOLUTION: The speech recognition device is provided with: a speech input part 11 for inputting speech uttered and an acoustic echo generated by a sound source; an echo cancel coefficient control unit 18 for controlling an echo cancel coefficient for canceling the acoustic echo; an echo cancel part 12 for using the audio signal from the sound source to cancel an acoustic echo component included in a signal from the speech input part and outputting it as a sound signal; a speech recognition part 13 for executing speech recognition processing on the basis of the speech signal from the echo cancel part; and a speech recognition performance determination part 14 for determining speech recognition performance on the basis of a result of recognition in the speech recognition part. The echo cancel coefficient control unit 18 returns the current echo cancel coefficient to that before the speech recognition performance decreases when it is determined by the speech recognition performance determination part that the speech recognition performance has lowered. COPYRIGHT: (C)2010,JPO&INPIT |