Analysis updated 2026-05-18
Turn a fixed-camera talking head video into a captioned 9:16 short automatically with Codex.
Have Codex install all needed tools step by step for a non-technical user.
Keep sensitive unreleased video content on a local disk instead of a syncing cloud folder.
| vizance/codex-talking-head-video-editor | aerdelan/housand-domaintoolmatrix | affaan-m/opencode | |
|---|---|---|---|
| Stars | 22 | 22 | 22 |
| Language | — | TypeScript | TypeScript |
| Last pushed | — | — | 2026-02-09 |
| Maintenance | — | — | Maintained |
| Setup difficulty | easy | easy | easy |
| Complexity | 2/5 | 2/5 | 2/5 |
| Audience | general | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Only works with a fixed-camera, single-shot, 16:9 talking head video on macOS or Windows, not for handheld, multi-camera, or interview footage.
Codex Talking Head Video Editor is a self-contained guide written for people with no technical background, designed to have Codex, an AI coding assistant, install everything needed and then automatically edit a specific kind of video. The source video must be a fixed-camera, single continuous shot, horizontal talking head recording, and the tool turns it into a vertical 9:16 short video with captions. The recommended way to use it is to download a file called GUIDE.md into a new empty folder, install and open the Codex app pointed at that folder, and ask Codex to read the guide fully and walk through the installation step by step, asking only one question at a time and automating whatever it can. Once installation is done, the user hands Codex a video and simply asks it to edit a new video. The README also mentions that a Codex able to read web pages can be given the GitHub URL of GUIDE.md directly, though downloading the file locally first and then handing it to Codex is described as the more reliable approach. There is a specific privacy warning: if the source video includes the presenter, a client, a company, or content that has not been made public yet, the working folder should be kept on a local disk rather than inside OneDrive, iCloud, Dropbox, NAS, or any other automatically syncing storage location. The README lists its supported scenarios narrowly: macOS on Apple Silicon or Intel, Windows 10 or 11 64-bit with native PowerShell, and source footage that is a fixed-camera, single continuous shot, 16:9 .mov or .mp4 talking head video. The default finished output uses a black background with white captions and a call to action inviting viewers to follow the creator's Instagram. This workflow is explicitly not meant for handheld selfie footage, walking shots, interviews, vlogs, or videos edited from multiple cameras. All the detailed installation, setup, operation, and troubleshooting steps live in the separate GUIDE.md file rather than in this README.
A Codex-driven guide that installs itself for non-technical users and automatically turns a fixed-camera talking head video into a captioned 9:16 short.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.