| CPC G10L 19/018 (2013.01) [G10L 25/18 (2013.01); G10L 25/21 (2013.01); G10L 25/51 (2013.01); G10L 25/84 (2013.01); G10L 25/90 (2013.01)] | 19 Claims |
|
1. A howling suppression method, performed by a computer device, the method comprising:
obtaining a current audio signal corresponding to a current time period, and performing frequency domain transformation on the current audio signal, to obtain a frequency domain audio signal;
dividing the frequency domain audio signal, to obtain subbands, and determining a target subband from the subbands;
obtaining a current howling detection result and a current voice detection result that correspond to the current audio signal, and determining a subband gain coefficient corresponding to the current audio signal based on the current howling detection result and the current voice detection result, the determined subband gain coefficient comprising a preset decreasing coefficient, a preset first increasing coefficient, or a preset second increasing coefficient, wherein the determined subband gain coefficient comprises the preset decreasing coefficient when the current voice detection result indicates that the current audio signal does not comprise a voice signal and the current howling detection result indicates that the current audio signal comprises a howling signal, the determined subband gain coefficient comprises the preset first increasing coefficient when the current voice detection result indicates that the current audio signal comprises the voice signal and the current howling detection result indicates that the current audio signal comprises the howling signal, and the determined subband gain coefficient comprises the preset second increasing coefficient when the current howling detection result indicates that the current audio signal does not comprise the howling signal, the preset first increasing coefficient and the preset second increasing coefficient are each associated with a respective speed of restoration to an initial subband gain value, and the preset first increasing coefficient is associated with a greater speed of restoration to the initial subband gain value than the preset second increasing coefficient;
obtaining a past subband gain corresponding to an audio signal within a past time period, and calculating a current subband gain corresponding to the current audio signal based on the subband gain coefficient and the past subband gain; and
suppressing howling on the target subband based on the current subband gain, to obtain a first target audio signal corresponding to the current time period,
wherein the obtaining the current howling detection result comprises extracting an initial audio feature based on an initial audio signal corresponding to the current time period; extracting a first past audio feature based on a first past audio signal corresponding to a first past time period, the first past time period being before the current time period; calculating a first similarity measure between the initial audio feature and the first past audio feature; and determining the current howling detection result based on the first similarity measure.
|