Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
VOICE SIGNAL PROCESSING METHOD AND DEVICE, TERMINAL, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/000597
Kind Code:
A1
Abstract:
A voice signal processing method and device, a terminal, and a storage medium. The method comprises: obtaining a compressed narrowband voice signal (S101); extracting a frequency domain feature of the narrowband voice signal (S102); inputting the frequency domain feature of the narrowband voice signal into a trained deep noise reduction self-encoder neural network model for nonlinear fitting, and obtaining a frequency domain feature of a full-band voice signal (S103); and converting the frequency domain feature of the full-band voice signal into a power spectrum of the full-band voice signal, performing inverse Fourier transformation on the power spectrum of the full-band voice signal by using phase information of a corresponding narrowband signal, and obtaining the full-band voice signal (S104). The compressed narrowband voice signal is subjected to bandwidth recovery by using the deep noise reduction self-encoder neural network model, thereby improving the quality and intelligibility of the voice signal.

Inventors:
CHEN FEI (CN)
YE FUQIANG (CN)
Application Number:
CN2020/078944
Publication Date:
January 07, 2021
Filing Date:
March 12, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV SOUTHERN SCI & TECH (CN)
International Classes:
G10L19/16; G10L25/24; G10L25/30
Foreign References:
CN110322891A2019-10-11
CN102870156A2013-01-09
CN107452389A2017-12-08
US20130332171A12013-12-12
Attorney, Agent or Firm:
BEYOND ATTORNEYS AT LAW (CN)
Download PDF: