AI video editor for talking-head clips: drop a raw video and walk it through six steps, each paused for a human check, to a finished video with styled captions and b-roll. WhisperX transcribes it with word-level timing into an editable transcript you can export as .txt, dead air can be cut automatically, and captions come in four styles (Classic, MrBeast, Hormozi, Clean) with an optional karaoke mode that pops each word as it is spoken. Claude reads the transcript and your project context, picks the words worth highlighting and plans a cutaway every 5 to 8 seconds, using your own reference images where they fit and Flux-generated images where they don't, and any plan can be adjusted with a plain-English instruction. FFmpeg composites the final render. Hosted on Cloud Run behind invite-only Firebase sign-in, with direct browser uploads to Cloud Storage, each user's own API keys remembered in Firestore, and a running cost history per session. Started as a fork of the open-source ButterCut project.
Tech Stack
- Python
- Streamlit
- WhisperX
- FFmpeg
- Claude API
- Flux
- Replicate
- Firebase Auth
- Firestore
- Cloud Run
- Cloud Storage
More Projects
Hostaluego
Prospecting engine for a dealer that sells and leases copiers and telecom to French businesses and public buyers. In that trade the only real buying trigger is a contract coming to an end, so the app hunts timing, not copy: it pulls companies from the French government's open company registry and spots dated moves like a head-office transfer, reads public procurement data (DECP) for copier and telecom contracts nearing their term, and scores every lead out of 100. Each qualified lead is checked on Google Places, and only the ones with a reachable French phone number are kept. On any lead, Claude writes a call brief, the "Déballe": why now, what is known with its source, an estimated monthly print volume and a ready phone opener, and it never mistakes a leasing company holding the contract for a competing dealer. Leads move into a built-in CRM with follow-ups, a daily call list, a planning view and a dashboard, and a call recorded on speakerphone is transcribed with Google Speech-to-Text and turned into a call report by Claude. Also: per-sector territories, a customer file import (CSV or Excel) that keeps existing clients out, an admin area with per-user daily lead limits and a read-only view-as mode, and a costs page that tracks every paid API call. Next.js on Firebase App Hosting with Firestore, invite-only sign-in, French UI.
Tahel Timer
A visual timer for young children, for iOS and Android. The child picks one of 12 timers (hourglass, clock, candle, race, ice cream, balloon, battery, pizza, sunset, juice, stars and rainbow) and watches the time run out, no reading needed, from 1 minute to 2 hours. Parents can add or take away a minute while it runs, lock it against little fingers and choose among 10 ringtones, and a running timer survives an app restart and still rings when the app is in the background or closed, through a local notification. Free with 4 timers, 3 ringtones and up to 30 minutes; a one-time Full version purchase, handled by RevenueCat behind a parental gate, unlocks the rest. 10 languages, no account, no ads, no tracking. First made for my daughter Tahel.
Picto Owl
A visual day schedule for children who need to see their day, like many children with autism or ADHD, and for the parents, teachers and therapists who set it up, for iOS and Android. Big picture cards show each step of the day: the child taps a card when it is done and always sees what comes now and next. 44 built-in pictures plus cards made from your own photos, a schedule for each child with their own photo, cards reordered by drag behind a hold-to-edit pencil, a first-then board for hard moments, and the day printed as a one-page PDF for the fridge or the classroom. Calm by design: soft colors, music off unless turned on, animations that follow the phone's reduce motion setting. Free with one schedule of up to 10 cards; a one-time Full version purchase, handled by RevenueCat behind a parental gate, unlocks the rest. 10 languages, no account, no ads, no tracking, and photos never leave the device.