AUDIO PROCESSING METHOD, APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

Title:

AUDIO PROCESSING METHOD, APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

Document Type and Number:

WIPO Patent Application WO/2021/082823

Kind Code:

A1

Abstract:

Provided are an audio processing method, apparatus, computer device (1200), and storage medium. The method is executed by the computer device (1200), and comprises: obtaining first audio data, the sampling rate of said first audio data being a first sampling rate (21); processing the first audio data by means of a super-resolution model to obtain an output result of the super-resolution model; the super-resolution model is a machine learning model obtained by taking a first audio sample as input data and a second audio sample as target data and performing deep learning training; the first audio sample being obtained by means of downsampling the second audio sample by a factor of N (22); obtaining the second audio data according to the output result of the super-resolution model, the sampling rate of the second audio data being a second sampling rate; the second sampling rate being N times the first sampling rate (23). The super-resolution model can more accurately restore the high-frequency components missing from the audio data of the low sampling rate, significantly improving audio processing performance.

Inventors:

CHEN FASHENG (CN)

Application Number:

PCT/CN2020/117317

Publication Date:

May 06, 2021

Filing Date:

September 24, 2020

Export Citation:

Click for automatic bibliography generation Help

Assignee:

TENCENT TECH SHENZHEN CO LTD (CN)

International Classes:

G10L21/003; G06N3/04; G06N3/08; G06N20/00; G10L21/007

Domestic Patent References:

WO2019063547A1	2019-04-04
WO2019197710A1	2019-10-17

Foreign References:

CN103606373A	2014-02-26
CN109147805A	2019-01-04
CN110797038A	2020-02-14
CN109147804A	2019-01-04
CN102610267A	2012-07-25
CN109074813A	2018-12-21

Other References: