Voice separation and paragraph affiliation method and system for teleconference scene
By constructing a logical map of claims-arguments and main lines-controversial points on the remote conference platform, the problem of difficulty in distinguishing the content of speakers' speeches in multi-party meetings was solved, and clear attribution and efficient editing of meeting records were achieved.
Patent Information
- Application Number
- CN202511003591.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-21
- Publication Date
- 2025-10-17
- Estimated Expiration
- 2045-07-21
AI Technical Summary
In remote conferences with multiple participants, existing technologies make it difficult to effectively distinguish speakers and the logical structure of their speeches, resulting in unclear attribution of meeting minutes and inefficient editing.
By matching the voice processing engine and temporarily accessing the target conference window on the conference platform, two-way voice separation processing is performed, and a logical map of the proposition-argument relationship and the main line-dispute point is constructed. Automated meeting records are generated based on the voice logical map, and storage management is carried out in combination with permission settings.
It realizes the two-way semantic logic separation and structured attribution of remote conference voice content, improving the accuracy of meeting records and the efficiency of automatic typesetting.