Loading patent…
AssigneeAtlassian US, Inc.
InventorsVictoria Tien Yu Chi, Erin Elizabeth Byerly, Avanika Krishnaswamy, Charles Todd Bracken, Matthew Job Granmoe, Sergio Pereira
Estimating the term…
The present disclosure provides a video editing system that receives a recorded video object configured to cause playback of a video recording on a client device. A transcript file object is generated based on the recorded video object. The system applies iterative multi-loop transcript refinement operations to the transcript file object, including inputting the transcript file object and text disfluencies deletion instructions to a large language model (LLM) to generate a first refined transcript file object, and inputting the first refined transcript file object and advanced text deletion instructions to the LLM to generate a second or subsequent refined transcript file object where multiple additional iterative multi-loop transcript refinement operations are deployed to further refine the transcript while mitigating unwanted transcript additions. The system generates video edit instructions based on the second (or subsequent) refined transcript object and generates an updated recorded video object based on the video edit instructions.
A video editing system comprising at least one processor and a memory storing instructions that are operable, when executed by the processor, to cause the video editing system to:
A computer-implemented method comprising:
A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor of a video editing system, cause the video editing system to perform operations comprising:
Dependencies are inferred from the claim text.