| 摘要 |
<p>Echo is canceled in a communication device including at least one microphone and a loudspeaker. Signals directed to the loudspeaker are filtered through at least one first filter. The filtered signals are subtracted from signals received by the microphone. At least one second filter filters the subtracted results to produce a signal corresponding to speech signals received by the microphone. At least one third filter filters the subtracted results to produce a signal corresponding to echo from the loudspeaker. The output from the third filter is subtracted from the output from the second filter to produce an echo-canceled signal. The device may include a number of microphones, and the first, second, and third filters may each include a number of filters equal to the number of microphones. The outputs from the second filter may be added, the outputs from the third filter may be added, and the sum of the outputs from the third filter may be subtracted from the sum of the outputs of the second filter. The output from the third filter may be weighted before being subtracted from output of the second filter.</p> |
