US 12,334,087 B2
Psychoacoustics-based audio encoding method and apparatus
Meng Wang, Shenzhen (CN); Zhuo Wang, Shenzhen (CN); and Fan Fan, Shenzhen (CN)
Assigned to Huawei Technologies Co., Ltd., Shenzhen (CN)
Filed by HUAWEI TECHNOLOGIES CO., LTD., Guangdong (CN)
Filed on Nov. 18, 2022, as Appl. No. 17/990,388.
Application 17/990,388 is a continuation of application No. PCT/CN2021/086256, filed on Apr. 9, 2021.
Claims priority of application No. 202010444294.2 (CN), filed on May 22, 2020.
Prior Publication US 2023/0091607 A1, Mar. 23, 2023
Int. Cl. G10L 15/22 (2006.01); G06F 3/16 (2006.01); G10L 19/02 (2013.01); G10L 25/21 (2013.01)
CPC G10L 19/0204 (2013.01) [G06F 3/165 (2013.01); G10L 25/21 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A psychoacoustic-based audio encoding method, comprising:
receiving audio data;
decoding the audio data;
obtaining auditory feature information of a user, wherein the auditory feature information comprises at least one of the following: personal information, listening test result information, or frequency response curve information;
calculating a psychoacoustics model parameter of the user based on the auditory feature information of the user, wherein the psychoacoustics model parameter comprises an intra-band masking parameter, a slope of a low-frequency inter-band masking line, a slope of a high-frequency inter-band masking line, and a human ear quiet threshold curve, and wherein the calculating comprises:
performing fitting on quiet threshold test result information of the user to obtain the human ear quiet threshold curve; and
performing fitting on masking test result information of the user to obtain the intra-band masking parameter, the slope of the low-frequency inter-band masking line, and the slope of the high-frequency inter-band masking line; and
encoding the decoded audio data based on the psychoacoustics model parameter of the user.