PURPOSE: To provide a speech recognition device in which the operability of a device can be improved by preventing the malfunction of the device due to a surrounding noise even at the time of using a hand-free microphone.
CONSTITUTION: When the position of a speaker 50 detected by a reflective sensor part 1 is proper to a hand-free microphone 2, a speaker monitoring part 3 outputs an instruction for validating the speech recognized result of a speech signal recognizing part 4 to a work station 6. The work station 6 inputs the instruction for validating the recognized result, and operates a processing based on the speech recognized result inputted from the voice signal recognizing part 4. Therefore, when the position of the speaker 50 is not proper to the hand-free microphone 2, the recognized result of the surrounding noise collected by the hand-free microphone 2 is not processed by the work station 6. Thus, it is possible to prevent the malfunction of the device even at the time of using the hand-free microphone 2 whose operability is satisfactory to the speaker 50.
MATSUURA HIROSHI
NITTA TSUNEO