Animation video frame intelligent selection dialogue generation system

Through the intelligent selection of dialogue generation system for animation video frames, computer vision and natural language processing technology are used to automatically recognize scenes and characters in animations and generate dialogues, which solves the problem of time-consuming and labor-intensive and incoherent dialogues, and improves the efficiency of animation production and Audience experience.

CN118918223AActive Publication Date: 2024-11-08BEIJING YOUQU TIME CULTURE TECH CO LTD

Patent Information

Application Number
CN202410921431.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-08
Estimated Expiration
2044-07-10

Smart Images

  • Figure CN118918223A_ABST
    Figure CN118918223A_ABST
Patent Text Reader

Abstract

The invention discloses an animation video frame intelligent selection dialogue generation system, and particularly relates to the technical field of computer vision, animation videos in a role library are converted into a video frame sequence, a computer vision technology is used for detecting a role appearing in each frame and tracking the role, appearance features of the detected role are extracted, and a dialogue is generated according to the extracted appearance features. Matching the extracted appearance characteristics of the role with the appearance characteristics of the role obtained from the role library, determining the identity of the detected role, establishing a role tone library which comprises audio data and marked tones, extracting audio characteristics from the audio data, carrying out tone recognition, and carrying out character recognition; the recognized timbre is matched with pre-labeled role timbre, dialogue content of each role is extracted according to the timbre recognition result, the dialogue content is converted into a text form, subject similarity judgment is conducted on the extracted dialogue content, the dialogue content with the similar subject is modified according to the judgment result, and the dialogue content with the similar subject is obtained. And generating new dialogue content.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Text-to-speech conversion method, device, electronic equipment and storage medium

    CN112765971A

  • Audio generation method and device of audio book and readable storage medium

    CN116665637A

  • Speech synthesis method, electronic equipment and storage medium

    CN116778900A

  • Contrast learning-based dialogue abstract method for specific role

    CN118277548A

  • Dialogue analysis system and dialogue analysis program

    JP2019028910A

Cited By

  • Video generation method and device, terminal and computer readable storage medium

    CN121000947A