Voice training partner method and device, electronic equipment and storage medium

By acquiring and processing the prompt words and models for voice training to generate emotional broadcast voices, the problems of traditional intelligent customer service training devices being labor-intensive and emotionally fixed are solved, and an efficient and personalized voice training experience is achieved.

CN120748407APending Publication Date: 2025-10-03AGRICULTURAL BANK OF CHINA
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202511070884.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-31
Publication Date
2025-10-03

AI Technical Summary

Technical Problem

Traditional intelligent customer service training devices rely on administrators to design a large number of speech templates and speech processes, which consumes huge manpower. In addition, the TTS model cannot specify the speaker's emotions, and the tone is always neutral, which has poor realism and experience, affecting the training effect.

Method used

By obtaining the common prompt words for voice training and the current prompt words for the simulated business scenario in the current dial test, the preset text generation model and speech generation model are used to generate the current broadcast voice with emotion. The preset speech recognition model is combined to process the user's response and dynamically adjust the content of the next round of dialogue.

Benefits of technology

There is no need for administrators to design speech templates, which reduces the cost of training, improves the efficiency and freedom of broadcast text generation, and the generated emotional voice enhances the sense of reality and experience, thereby improving the training effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120748407A_ABST
    Figure CN120748407A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a voice partner training method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a universal cue word of voice partner training and a current cue word corresponding to a current round of dialogue in a simulation service scene in a current dial test; the current broadcast text is accurately and conveniently generated based on the general prompt word, the current prompt word and the preset text generation model, so that the generation efficiency of the current broadcast text is improved; if it is detected that the current broadcast text does not contain the last sentence identifier, determining a current broadcast voice corresponding to the current round of dialogue based on the current broadcast text, the current prompt word and a preset voice generation model; the current broadcast voice is the voice with emotions, so that the emotional effect of voice broadcast is realized; determining a current response text corresponding to the current round of dialogue based on the received current response voice of the user and a preset voice recognition model; and based on the current broadcast text, the current response text and the current cue word, determining a next cue word corresponding to the next round of dialogue.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Large model training method and device based on voice task, medium and equipment

    CN121768368A