| 摘要 |
PURPOSE:To perform speech recognition processes by using signals from a subtracting function so as to reduce the erroneous recognition rate by providing a noise microphone which receives ambient noise signals and the subtracting function which subtraction- processes the signal of the noise microphone from the signal of an array microphone. CONSTITUTION:An array microphone 1 is composed of plural microphones arranged near an uttering person in the form of an array and mainly receives and synthesizes sound signals. A noise microphone 2 which mainly receives ambient noises is provided beside the array microphone 1 apart from uttering person. When the distance (d) between the microphones 1 and 2 is made sufficiently short, sounds can be regarded as plane waves and the microphones 1 and 2 are arranged in such a way that they can become parallel with the plane waves. Signals of the microphones 1 and 2 are sent to a subtractor 3 which subtracts the output of the noise microphone 2 from that of the array microphone 1. On the downstream side of the subtractor 3, a speech recognizing device 4 is provided and the device 4 recognizes the signal of the subtractor 3. Outputs of the microphones 1 and 2 are subtracted against low-frequency noises and a result of the subtraction is inputted to the speech recognizing device. |
