| 摘要 |
Provided is an echo canceller device capable of delivering stable echo cancellation performance even immediately after utterance from a speaker. An echo canceller device (10) is provided with a first filter unit (111) that generates a first output signal y1(k) by filtering a first input signal x1(k) with a filter coefficient h(k), an adaptation unit (13) that modifies the filter coefficient h(k) on the basis of an adaptation algorithm so that a first error signal e1(k) becomes small, a first delaying unit (141) that delays a first objective signal d1(k) by a certain time period τ and outputs the signal as a second objective signal d2(k), a second delaying unit (142) that delays the first input signal x1(k) by the certain time period τ and outputs the signal as a second input signal x2(k), and a second filter unit that generates a second output signal y2(k) by filtering the second input signal x2(k) by the filter coefficient h(k) modified by the adaptation unit (13). |
