Title:
MODEL TRAINING METHOD AND APPARATUS, SPEECH-TO-SPEECH TRANSLATION METHOD AND APPARATUS, AND MEDIUM
Document Type and Number:
WIPO Patent Application WO/2023/207638
Kind Code:
A1
Abstract:
A model training method and apparatus, a speech-to-speech translation method and apparatus, and a medium. The model training method comprises: acquiring a speech recognition sample and a real speech-to-speech translation sample (S310); generating a pseudo-labeled speech-to-speech translation sample according to the speech recognition sample (S320); and training a speech-to-speech translation model according to the pseudo-labeled speech-to-speech translation sample and the real speech-to-speech translation sample (S330). The model training method can solve the problem of low model training precision caused by lack of translation sample data.
Inventors:
DONG QIANQIAN (CN)
YUE FENGPENG (CN)
KO YU TING (CN)
WANG MINGXUAN (CN)
BAI QIBING (CN)
YUE FENGPENG (CN)
KO YU TING (CN)
WANG MINGXUAN (CN)
BAI QIBING (CN)
Application Number:
PCT/CN2023/088492
Publication Date:
November 02, 2023
Filing Date:
April 14, 2023
Export Citation:
Assignee:
BEIJING YOUZHUJU NETWORK TECH CO LTD (CN)
International Classes:
G10L13/02; G10L15/06
Foreign References:
| CN114822499A | 2022-07-29 | |||
| CN111738025A | 2020-10-02 | |||
| CN112966529A | 2021-06-15 |
Other References:
JIA YE; JOHNSON MELVIN; MACHEREY WOLFGANG; WEISS RON J.; CAO YUAN; CHIU CHUNG-CHENG; ARI NAVEEN; LAURENZO STELLA; WU YONGHUI: "Leveraging Weakly Supervised Data to Improve End-to-end Speech-to-text Translation", ICASSP 2019 - 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), IEEE, 12 May 2019 (2019-05-12), pages 7180 - 7184, XP033565891, DOI: 10.1109/ICASSP.2019.8683343
YE JIA; MICHELLE TADMOR RAMANOVICH; TAL REMEZ; ROI POMERANTZ: "Translatotron 2: Robust direct speech-to-speech translation", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 3 December 2021 (2021-12-03), 201 Olin Library Cornell University Ithaca, NY 14853, XP091110046
YE JIA; MICHELLE TADMOR RAMANOVICH; TAL REMEZ; ROI POMERANTZ: "Translatotron 2: Robust direct speech-to-speech translation", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 3 December 2021 (2021-12-03), 201 Olin Library Cornell University Ithaca, NY 14853, XP091110046
Attorney, Agent or Firm:
ZHIFAN & PARTNERS (CN)
Download PDF:
Previous Patent: COMMUNICATION METHOD AND APPARATUS
Next Patent: CONTROL MECHANISM OF SURGICAL INSTRUMENT AND SURGICAL ROBOT
Next Patent: CONTROL MECHANISM OF SURGICAL INSTRUMENT AND SURGICAL ROBOT