The invention discloses a voice interaction method, a server and a computer readable storage medium, and the method comprises the steps: receiving a current voice request forwarded by a vehicle, carrying out the retrieval of a knowledge database according to the current voice request, obtaining a first retrieval result, and obtaining a second retrieval result based on a large language model; according to the prompt information template and the first retrieval result, slot recognition, application program interface prediction and application program interface parameter filling are carried out on the current voice request to obtain an execution result of application program interface parameter filling, and the execution result is output and issued to the vehicle to complete voice interaction. Thus, in the model reasoning process, the knowledge database is retrieved according to the current voice request, so that the model can complete reasoning of the current voice request based on the retrieval result and the prompt information template, and the timeliness of the reasoning result can be related to information in the knowledge database; the condition that the model gives out expired answers due to expired training data of the model is avoided, and the reasoning precision of the model is guaranteed.

    本申请公开一种语音交互方法、服务器及计算机可读存储介质,所述方法包括:接收到车辆转发的当前语音请求,根据当前语音请求对知识数据库进行检索,得到第一检索结果,基于大语言模型,根据提示信息模板和第一检索结果对当前语音请求进行槽位识别和应用程序接口预测以及应用程序接口参数填充以得到应用程序接口参数填充的执行结果,输出执行结果下发至车辆以完成语音交互。如此,本申请在模型推理过程中以当前语音请求检索知识数据库,使得模型可基于检索结果和提示信息模板,完成当前语音请求的推理,推理结果的时效性可与知识数据库中的信息相关,避免模型的训练数据过期而导致模型给出过期的答案的情况,保障模型的推理精度。


    Access

    Download


    Export, share and cite



    Title :

    Voice interaction method, server and computer readable storage medium


    Additional title:

    语音交互方法、服务器及计算机可读存储介质


    Contributors:
    DING PENGJIE (author) / ZHAO QUN (author) / NING HONGKE (author) / FAN JUNFENG (author) / ZHU QIYU (author)

    Publication date :

    2024-05-14


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G10L Analyse oder Synthese von Sprache , SPEECH ANALYSIS OR SYNTHESIS / B60R Fahrzeuge, Fahrzeugausstattung oder Fahrzeugteile, soweit nicht anderweitig vorgesehen , VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR



    Voice interaction method, server and computer readable storage medium

    ZHU QIYU / FAN JUNFENG / ZHAO QUN | European Patent Office | 2024

    Free access

    Voice interaction method, server and computer readable storage medium

    FAN JUNFENG / ZHAO QUN / DING PENGJIE et al. | European Patent Office | 2024

    Free access

    Voice interaction method, server and computer readable storage medium

    FAN JUNFENG / DING PENGJIE / ZHAO QUN et al. | European Patent Office | 2023

    Free access

    Voice interaction method, server and computer readable storage medium

    FAN JUNFENG / ZHAO QUN / JIAN TAOWEN et al. | European Patent Office | 2024

    Free access

    Voice interaction method, server and computer readable storage medium

    ZHU QIYU / NING HONGKE / FAN JUNFENG et al. | European Patent Office | 2024

    Free access