A walkthrough of a single real run: thirty minutes of raw interview footage in, a set of finished vertical clips out, with the actual time each stage takes.
Upload and analysis
A thirty-minute source file at 1080p is typically two to four gigabytes. Upload time depends entirely on your connection and is usually the longest single stage of the process.
Analysis — transcription, prosody, speaker detection, segment identification — runs in well under a minute for a source of this length, because the three audio passes run concurrently.
Selection
The segment list comes back ranked, with the transcript excerpt for each so you can judge them by reading rather than by scrubbing. A thirty-minute interview generally yields ten to fourteen viable segments.
This is the stage worth spending time on. Reading fourteen excerpts takes two or three minutes and determines the quality of everything downstream.
Generation and export
Selected segments render in parallel. Ten clips at 4K vertical typically complete in one to two minutes total rather than sequentially.
Export presets handle the aspect ratio and bitrate differences between TikTok, Reels and Shorts. The same edit is re-rendered per target rather than uploaded once and letting each platform re-encode it.
- Upload: connection-bound, usually the longest stage
- Analysis: under a minute for a 30-minute source
- Selection: 2–3 minutes of reading, the stage that matters most
- Render: 1–2 minutes for ten 4K vertical clips
The honest total
Excluding upload, the working time is roughly five minutes, most of it spent choosing segments. The "under two minutes" figure refers to generation and render, not to the end-to-end session, and we would rather state that plainly than have you discover it on your first run.
Try it on your own footage
Thirty processing minutes a month on the free plan. No credit card required.
Get early access