US 12,283,285 B2
Automatic gain control method and device, and readable storage medium
Xiaoliang Chen, Beijing (CN); and Dahang Feng, Beijing (CN)
Assigned to SOUNDAI TECHNOLOGY CO., LTD., Beijing (CN)
Appl. No. 17/606,950
Filed by SOUNDAI TECHNOLOGY CO., LTD., Beijing (CN)
PCT Filed Oct. 31, 2019, PCT No. PCT/CN2019/114764
§ 371(c)(1), (2) Date Oct. 27, 2021,
PCT Pub. No. WO2020/220625, PCT Pub. Date Nov. 5, 2020.
Claims priority of application No. 201910358510.9 (CN), filed on Apr. 29, 2019.
Prior Publication US 2022/0215855 A1, Jul. 7, 2022
Int. Cl. G10L 21/034 (2013.01); G10L 21/0208 (2013.01); G10L 21/0216 (2013.01); G10L 21/0224 (2013.01); G10L 25/21 (2013.01); G10L 25/78 (2013.01)
CPC G10L 21/034 (2013.01) [G10L 21/0224 (2013.01); G10L 25/21 (2013.01); G10L 25/78 (2013.01); G10L 2021/02082 (2013.01); G10L 2021/02166 (2013.01); G10L 2025/783 (2013.01)] 17 Claims
OG exemplary drawing
 
1. An automatic gain control method, comprising:
for a far-field speech signal of a current frame, distinguishing between a target signal and a non-target signal;
according to a result of the distinguishing between the target signal and the non-target signal, determining a gain table calculation parameter of the far-field speech signal of the current frame, and obtaining a gain variation of the far-field speech signal of the current frame relative to a previous frame;
determining a gain value for the far-field speech signal of the current frame according to the gain variation; and
processing the far-field speech signal of the current frame according to the gain value determined, to obtain a processed speech signal;
wherein according to the result of the distinguishing between the target signal and the non-target signal, determining the gain table calculation parameter of the far-field speech signal of the current frame, and obtaining the gain variation of the far-field speech signal of the current frame relative to the previous frame, comprises:
according to the result of the distinguishing between the target signal and the non-target signal, determining the gain table calculation parameter of the far-field speech signal of the current frame;
obtaining a gain of the previous frame and a smoothing coefficient;
calculating a gain of the far-field speech signal of the current frame, according to an equation: gain_cur(t)=α*gain_cur(t−1)+(1−α)*gain, based on the gain table calculation parameter, the gain of the previous frame, and the smoothing coefficient; and
obtaining the gain variation of the far-field speech signal of the current frame relative to the previous frame, according to an equation Δgain=gain_cur(t)−gain_cur(t−1), based on the gain of the previous frame and the gain of the far-field speech signal of the current frame,
where t is a count of frames, a is the smoothing coefficient, gain_cur(t−1) is the gain of the previous frame, gain_cur(t) is the gain of the far-field speech signal of the current frame, Δgain is the gain variation, and gain is the gain table calculation parameter of the far-field speech signal of the current frame;
wherein determining the gain value for the far-field speech signal of the current frame according to the gain variation, comprises:
in a case where the gain variation is greater than a predetermined threshold, determining the gain value for the far-field speech signal of the current frame according to a gain table;
otherwise in a case where the gain variation is not greater than the predetermined threshold, using a gain value of the previous frame as the gain value for the far-field speech signal of the current frame.