| 主权项 |
1. A method for filtering out speech interference, applicable to an electronic device which comprises a processor and a microphone, the method comprising:
defining a time threshold based on a probability distribution model by the processor; when a current instruction is recognized, by the processor, from a speech input received by the microphone, obtaining a reference instruction recognized from the speech input by the processor, wherein the current instruction is recognized right after the reference instruction, the reference instruction and the current instruction correspond to a first time point and a second time point respectively; determining whether speech interference occurs according to a comparison result and a status corresponding to the first time point by the processor, wherein the comparison result is generated by comparing the time threshold with a time interval between the first time point and the second time point, and the status corresponding to the first time point being set based on a recognition status of the speech input recognized at the first time point, wherein the status is one of a plurality of statuses, and the statuses at least comprise a first status, a second status, and a third status; filtering out the reference instruction and the current instruction by the processor if the speech interference occurs; and outputting one of the reference instruction and the current instruction by the processor if the speech interference does not occurs. |
