PURPOSE: To obtain a sufficiently high speech recognition ratio by using a signal transmitted from a present device as a semi-signal for correction.
CONSTITUTION: This speech recognition/response device is provided with a reference signal sending means 7 for sending a reference signal to be the reference of gain adjustment, a means 4 for calculating gain for correcting the level of an input signal based upon the level difference between the reference signal and an input signal and an amplifier means 6 for amplifying the input signal in accordance with the calculated gain value. For instance, a signal is originated from a specific line of the present device to another line in the present device at the time of application of a power supply e.g. to connect both lines, the folding state of a signal utilizing a telephone circuit network is set up, a reference signal is sent, and the gain of the amplifier means 6 is automatically controlled in accordance with a level difference between the reference signal and its feedback input. Consequently, the attenuation of a voice signal due to the telephone circuit or the like can be accurately corrected and sufficiently high speech recognition ratio can be obtained.
JP2001147696 | SIGNAL PROCESSING SYSTEM |
JPS6121527 | VOICE RECOGNITION INPUT DEVICE |
JP7241005 | Head-up display system and moving object |