Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
AUDIO DATA PROCESSING METHOD AND APPARATUS, DEVICE, STORAGE MEDIUM, AND PROGRAM PRODUCT
Document Type and Number:
WIPO Patent Application WO/2024/055751
Kind Code:
A1
Abstract:
An audio data processing method and apparatus, a device, a storage medium, and a program product. The method comprises: acquiring a target audio data frame and K historical audio data frames associated with raw audio data (S101); on the basis of acquired N target cepstral coefficients of the target audio data frame, acquiring M first-order time derivatives and M second-order time derivatives (S102); acquiring N historical cepstral coefficients corresponding to each historical audio data frame, and on the basis of the acquired K*N historical cepstral coefficients, determining spectral dynamic features (S103); and inputting the N target cepstral coefficients, the M first-order time derivatives, the M second-order time derivatives and the spectral dynamic features into a target mask estimation model to obtain a target mask corresponding to the target audio data frame (S104), wherein the target mask is used for suppressing noise data in the raw audio data to obtain enhanced audio data corresponding to the raw audio data.

Inventors:
HUANG DAIYU (CN)
BAO FENG (CN)
LI YUEPENG (CN)
Application Number:
PCT/CN2023/108796
Publication Date:
March 21, 2024
Filing Date:
July 24, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
TENCENT TECH SHENZHEN CO LTD (CN)
International Classes:
G10L21/0208
Foreign References:
CN113823313A2021-12-21
CN114203163A2022-03-18
CN113963715A2022-01-21
CN1737906A2006-02-22
CN113870893A2021-12-31
US20190318755A12019-10-17
JP2006349840A2006-12-28
Attorney, Agent or Firm:
DEQI INTELLECTUAL PROPERTY LAW CORPORATION (CN)
Download PDF:
View/Download PDF      PDF Help



 
Previous Patent: COMMUNICATION DEVICE

Next Patent: SPEECH SYNTHESIS MODEL TRAINING METHOD, SPEECH SYNTHESIS METHOD, AND RELATED APPARATUSES