公开(公告)号：US12087289B2

公开(公告)日：2024-09-10

申请号：US17539005

申请日：2021-11-30

Applicant: HUAWEI TECHNOLOGIES CO.,LTD.

Inventor： Weiran Nie ,  Fuliang Weng ,  Youjia Huang ,  Hai Yu ,  Shumang Hu

IPC: G10L15/18 ,  G10L15/187 ,  G10L15/22 ,  G10L15/06 ,  G10L15/065 ,  G10L15/07 ,  G10L15/08

CPC classification number: G10L15/1822 ,  G10L15/18 ,  G10L15/187 ,  G10L15/22 ,  G10L15/063 ,  G10L15/065 ,  G10L15/07 ,  G10L2015/088

Abstract: A speech recognition method, apparatus, and device, and a computer-readable storage medium provided pertain to the field of artificial intelligence technologies. The method includes: obtaining or generating a dynamic target language model based on reply information of a first intent, where the dynamic target language model includes a front-end part and a core part; obtaining a speech signal, parsing the speech signal to generate a key word; and invoking the dynamic target language model to determine a second intent and a service content. The front-end part of the dynamic target language model parses out the second intent based on the key word, and the core part of the dynamic target language model parses out the service content based on the key word. The speech recognition method prevents a provided service content from deviating from a user requirement and achieves a good recognition effect.

公开(公告)号：US20220093087A1

公开(公告)日：2022-03-24

申请号：US17539005

申请日：2021-11-30

Applicant: HUAWEI TECHNOLOGIES CO.,LTD.

Inventor： Weiran Nie ,  Fuliang Weng ,  Youjia Huang ,  Hai Yu ,  Shumang Hu

IPC: G10L15/18 ,  G10L15/187

Abstract: A speech recognition method, apparatus, and device, and a computer-readable storage medium provided pertain to the field of artificial intelligence technologies. The method includes: obtaining or generating a dynamic target language model based on reply information of a first intent, where the dynamic target language model includes a front-end part and a core part; obtaining a speech signal, parsing the speech signal to generate a key word; and invoking the dynamic target language model to determine a second intent and a service content. The front-end part of the dynamic target language model parses out the second intent based on the key word, and the core part of the dynamic target language model parses out the service content based on the key word. The speech recognition method prevents a provided service content from deviating from a user requirement and achieves a good recognition effect.