Title:
SPEECH TEXT GENERATION METHOD AND APPARATUS, AND TRAINING METHOD AND APPARATUS FOR SPEECH TEXT GENERATION MODEL
Document Type and Number:
WIPO Patent Application WO/2024/077906
Kind Code:
A1
Abstract:
Provided in the present disclosure are a speech text generation method, which can be applied to the technical field of artificial intelligence and the field of intelligent customer service. The speech text generation method comprises: performing part-of-speech tagging on standard text, so as to obtain a part-of-speech tagging result; according to a modal particle distribution feature, determining a target part-of-speech from the part-of-speech tagging result; determining a predicted insertion position according to the position, in the standard text, of content corresponding to the target part-of-speech; inserting a target modal particle into the standard text according to the predicted insertion position, so as to obtain target spoken text; and generating target speech text according to the target spoken text. Further provided in the present disclosure are a training method for a speech text generation model, and a speech text generation apparatus, a training apparatus for a speech text generation model, and a device, a medium and a program product.
Inventors:
FENG MINGCHAO (CN)
CHEN MENG (CN)
QIN JIE (CN)
CHEN MENG (CN)
QIN JIE (CN)
Application Number:
PCT/CN2023/087793
Publication Date:
April 18, 2024
Filing Date:
April 12, 2023
Export Citation:
Assignee:
JINGDONG TECH INFORMATION TECH CO LTD (CN)
International Classes:
G10L15/26; G10L15/06; G10L15/18; G10L15/22
Foreign References:
CN115620726A | 2023-01-17 | |||
CN114912448A | 2022-08-16 | |||
CN114708868A | 2022-07-05 | |||
CN114218424A | 2022-03-22 | |||
CN108170674A | 2018-06-15 | |||
US10599767B1 | 2020-03-24 | |||
US20210312124A1 | 2021-10-07 |
Attorney, Agent or Firm:
CHINA SCIENCE PATENT & TRADEMARK AGENT LTD. (CN)
Download PDF: