US 12,308,015 B2
Training method and apparatus for speech translation model, speech translation method and apparatus, and device
Zhicheng Liu, Beijing (CN); Jian Tong, Beijing (CN); Chengqi Zhao, Beijing (CN); Tao Wang, Beijing (CN); and Mingxuan Wang, Beijing (CN)
Assigned to Beijing Youzhuju Network Technology Co., Ltd., Beijing (CN)
Filed by Beijing Youzhuju Network Technology Co., Ltd., Beijing (CN)
Filed on Jun. 24, 2024, as Appl. No. 18/752,273.
Application 18/752,273 is a continuation of application No. PCT/CN2023/082944, filed on Mar. 22, 2023.
Claims priority of application No. 202210319911.5 (CN), filed on Mar. 29, 2022.
Prior Publication US 2024/0347044 A1, Oct. 17, 2024
Int. Cl. G10L 15/06 (2013.01); G06F 40/58 (2020.01); G10L 15/18 (2013.01)
CPC G10L 15/063 (2013.01) [G06F 40/58 (2020.01); G10L 15/1815 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A training method for a speech translation model, comprising:
after entering a model training phase, controlling the speech translation model to execute a speech translation training task;
controlling the speech translation model to simultaneously execute an auxiliary training task of the speech translation training task; and
adjusting network parameters of the speech translation model according to the speech translation training task and the auxiliary training task, so as to obtain an updated speech translation model after training, so that a translation that is generated by the updated speech translation model and displayed to a user is real-time consistent with a speech to be translated,
wherein the speech translation training task is used for translating an original text participating in speech recognition, and the auxiliary training task is used for weakening a translation bias generated on the speech translation model caused by a bias between a text obtained with a speech recognition bias and text semantics.