No designer. No editor. No problem.
A raw selfie video becomes a polished, captioned, on-brand 9:16 film — word-accurate captions, auto-trimmed cuts, and a one-click branded render, all without anyone opening an editing app.
A blade-tool timeline that keeps captions glued
Make frame-accurate cuts anywhere in the clip — and word-level Whisper transcription keeps every caption glued through each trim, so the words never drift out of sync. Silence-aware sync and automatic filler-word cleanup tidy the “ums” and dead pauses for you.
- Word-level captions with silence-aware sync
- Automatic filler-word cleanup — “ums” and dead air tidied
- Frame-accurate blade cuts; captions stay perfectly in sync
Captions stay glued through every cut
An auto-tightened first cut, waiting for you
A first-draft cut arrives already trimmed — dead air shaved off the front and back — so you start from a clean, watchable take instead of raw footage. Transcription is warmed the moment a response is submitted, so the words are already waiting; you never sit watching a spinner.
- Dead air shaved off the head and tail automatically
- Captions ready before you open it — no spinner
- Fully editable if you want the final say
Fully editable — you keep the final say
Press render; watch a branded film build itself
Press render and dedicated cloud infrastructure assembles the whole thing: an animated intro, your logos and palette, question cards, word-by-word caption pills, and a custom end-card — a finished 9:16 video, no editing app required. A live, self-healing progress view shows exactly where it’s up to.
- Animated intro, brand palette, caption pills & end-card
- Live, self-healing progress — no guessing, no refreshing
- A finished 9:16 video, no editing app required
It leads with a genuine smile — not a mid-blink frame
An on-device AI smile-picker leads each testimonial with a real smile instead of a mid-blink frame, so your thumbnails sell the story at zero API cost. AI sentiment tinting and cross-response Nest Insights surface the strongest moments, with NZ-English spelling out of the box and a te-reo-Māori-ready transcription path.
- On-device smile-picker — zero API cost
- Sentiment tinting & cross-response Nest Insights
- NZ-English spelling, te-reo-Māori-ready transcription
Picked on-device with TensorFlow.js — no cloud vision API, no per-use cost.
Blade cuts are physically reproduced at render — each kept range trimmed and stream-copy concatenated, verified frame- and sample-accurate, with a fallback so renders never hard-fail.
Renders retry only the render step on a known startup flake, while real errors fail fast.
The render worker is the source of truth — it writes the final video URL itself, so a dropped tab mid-render still delivers your film.
The smile-thumbnail runs entirely on-device with TensorFlow.js — no cloud vision API, no per-use cost.
Your customers record. We produce.
Collect a story and watch it come back as a finished, on-brand film.