Local-first tool for turning long YouTube videos into ready-to-post vertical clips with transcription, burned-in subtitles, and smart crop options.

• Download a single YouTube video with yt-dlp.
• Transcribe locally with faster-whisper.
• Score transcript windows for clip candidates.
• Export vertical 9:16 MP4 clips with SRT files.
• Burn subtitles into clips by default.
• Crop center or shift crop toward detected faces/people.
• Manage jobs and generated clips from a Next.js UI.
• Run locally with Python/Node or with Docker Compose.

• Python 3.12+
• Node.js 22+
• npm
• Network access for YouTube downloads and model downloads
• Enough CPU, disk, and time for transcription and video encoding

Docker users only need Docker and Docker Compose.

Copy the Docker env example:

Copy-Item .env.docker.example .env

For local Docker usage, defaults are enough:

FRONTEND_PORT=3000
BACKEND_PORT=8010
NEXT_PUBLIC_API_BASE=http://localhost:8010

Build and run:

docker compose --env-file .env up -d --build

Open:

frontend: http://localhost:3000
backend:  http://localhost:8010

Persistent local data:

backend/outputs -> /app/outputs
backend/jobs.json -> /app/jobs.json

Start backend:

cd backend
py -m venv .venv
.\.venv\Scripts\python.exe -m pip install -r requirements.txt
.\.venv\Scripts\python.exe -m uvicorn api:app --host 127.0.0.1 --port 8010

Start frontend in another terminal:

cd frontend
npm install
npm run dev

Open http://127.0.0.1:3000.

The backend can also run without the UI:

cd backend
.\.venv\Scripts\python.exe clipper.py "https://www.youtube.com/watch?v=..." --top 5 --min 35 --max 180

Quick test on the first 180 seconds:

.\.venv\Scripts\python.exe clipper.py "https://www.youtube.com/watch?v=..." --model Systran/faster-whisper-base --analyze-seconds 180 --top 1

Outputs are written under backend/outputs/.

GET    /api/health
POST   /api/jobs
GET    /api/jobs
GET    /api/jobs/{job_id}
DELETE /api/jobs
GET    /outputs/<generated-file>

For Docker/server deployments, NEXT_PUBLIC_API_BASE must be the browser-accessible backend URL:

NEXT_PUBLIC_API_BASE=https://api.example.com

The frontend also uses BACKEND_API_BASE internally for proxying API requests in Docker:

BACKEND_API_BASE=http://backend:8010

ClipForge is intended for local workflows and content you are allowed to process. Make sure you have rights or permission to download, transform, and republish source videos. Follow YouTube terms and applicable copyright law.

Do not expose the backend publicly without authentication, rate limits, request validation, quotas, and cleanup. The backend accepts URLs and runs expensive jobs.

backend/
  api.py                 FastAPI job API
  clipper.py             download/transcribe/score/export pipeline
  models/                optional crop detection model
  outputs/               generated local files, ignored by git
frontend/
  app/                   Next.js app router UI
  lib/apiClient.ts       API client helpers
  types/clip.type.ts     shared frontend types
docker-compose.yml       local two-service stack

MIT. See LICENSE.

Third-party notices live in NOTICE.

Created by mallexibra.