Inspiration
Our camera rolls are full of moments worth remembering. Too often, they stay there because editing takes more time than filming. We built Vlogii to turn everyday clips into a story people actually want to watch and share.
What it does
Vlogii turns raw footage into an editable vlog with AI-selected moments, captions, music, and beat-driven cuts.
Upload your clips, generate a first cut, and make it yours. Trim moments, adjust speed, change the look, and create a cover—all with preview, undo, and export.
How we built it
We paired Gemini’s video understanding with a Python and FFmpeg rendering pipeline, wrapped in a React interface. AI proposes the edit; code validates the timeline and renders the film. Beat analysis helps align cuts with music.
Challenges we ran into
The hardest part was turning creative AI suggestions into reliable edits: keeping clips, captions, and audio aligned while handling different video formats. We also made retries and undo preserve the last successful film.
Accomplishments we’re proud of
We built a working local MVP that takes footage from upload to an exported MP4, with music-driven styles, editable chapters, and custom covers.
The first cut is a starting point. The creator stays in control.
What we learned
A compelling first cut gets people started. Clear previews, easy adjustments, and dependable undo help them finish.
What’s next for Vlogii
Better storytelling, more visual styles, and richer natural-language editing—so creators can spend less time managing a timeline and more time sharing their lives.
Built With
- fastapi
- ffmpeg
- google-gemini-api
- html
- librosa
- pillow
- pydantic
- python
- react
- typescript
- vite
Log in or sign up for Devpost to join the conversation.