Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD, DEVICE, AND MEDIUM FOR SPEECH CONVERSION, FILE GENERATION, BROADCASTING, AND VOICE PROCESSING
Document Type and Number:
WIPO Patent Application WO/2021/083071
Kind Code:
A1
Abstract:
Provided are a method, device, and medium for voice conversion, file generation, broadcasting, and voice processing. During a speech conversion process, acoustic features are combined with pronunciation information, acoustic features are mapped to pronunciation information in at least one language, and speech conversion from a first sound source to the second sound source is completed by combining with a feature conversion relationship, learned in advance, of the pronunciation information to vocoder features; pronunciation information having weaker relevance to the language of the first sound source is used for speech conversion, the conversion result is less affected by the first sound source, and voice conversion quality is higher; in addition, using pronunciation information in at least one language, it is possible to expand the scope of language application of the first sound source, improving the intelligence of speech conversion.

Inventors:
ZHAO SHENGKUI (CN)
Application Number:
PCT/CN2020/123593
Publication Date:
May 06, 2021
Filing Date:
October 26, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ALIBABA GROUP HOLDING LTD (CN)
International Classes:
G10L15/02; G10L13/08; G10L19/16; G10L21/003
Foreign References:
CN110970014A2020-04-07
CN109377986A2019-02-22
CN108682426A2018-10-19
CN110111771A2019-08-09
US9558733B12017-01-31
CN109948124A2019-06-28
Attorney, Agent or Firm:
BEIJING SANYOU INTELLECTUAL PROPERTY AGENCY LTD. (CN)
Download PDF: