How sahandghavidel/ai-video-editor works
Ultimate Video Editor is a self-hosted, scene-based AI video production workspace. Users configure a pipeline, work with source videos and scenes, generate or refine narration and media, then assemble and export publish-ready assets. The graph combines the supplied route excerpts with README-documented capabilities; unsampled subsystems are represented at the existing API directory level, and their internal wiring is not inferred.
Main entry points: generate-tts-omnivoice/route.ts Narration TTS · add-image-overlay/route.ts Image Overlays · api Production Dashboard · separate-scene/route.ts Separate Scene
Components 21
Workspace
- Production Dashboard
src/app/api - Pipeline Config
src/components/PipelineConfig.tsx - Projects & Scenes
src/app/api
Scene Production
- Script & Prompts
src/app/api - Generate Scenes
src/app/api/generate-scenes/route.ts - Separate Scene
src/app/api/separate-scene/route.ts - Scene Media
src/app/api
Speech & Language
- Narration TTS
src/app/api/generate-tts-omnivoice/route.ts - English Dubbing
src/app/api/create-dubbed-en/route.ts - Persian Dubbing
src/app/api/create-dubbed-fa/route.ts - Subtitle Retrieval
src/app/api/download-youtube-subtitles/route.ts
Media & Delivery
- Video & Audio Edit
src/app/api - Image Overlays
src/app/api/add-image-overlay/route.ts - Upscale & Interpolate
REAL-Video-Enhancer/backend/src/RenderVideo.py - Metadata & Export
src/app/api
External Services
- Baserow
- MinIO
- FFmpeg
- AI & Speech Providers
- YouTube / yt-dlp
Other components
- Producer
Connections 25
- Producer → Production Dashboard: uses
- Production Dashboard → Pipeline Config: configures
- Production Dashboard → Projects & Scenes: manages
- Pipeline Config → Script & Prompts: enables steps
- Projects & Scenes → Script & Prompts: provides project
- Script & Prompts → Generate Scenes: creates scenes
- Generate Scenes → Baserow: reads and writes
- Separate Scene → Baserow: reads and updates
- Separate Scene → FFmpeg: probes duration
- Scene Media → Narration TTS: requests narration
- Narration TTS → AI & Speech Providers: generates speech
- Narration TTS → MinIO: uploads audio
- Narration TTS → Baserow: reads scenes
- English Dubbing → Baserow: reads and updates
- English Dubbing → MinIO: uploads audio
- Persian Dubbing → Baserow: reads and updates
- Persian Dubbing → MinIO: uploads assets
- Subtitle Retrieval → YouTube / yt-dlp: runs yt-dlp
- Video & Audio Edit → FFmpeg: processes media
- Image Overlays → FFmpeg: renders overlay
- Image Overlays → MinIO: uploads video
- Image Overlays → Baserow: updates scene
- Upscale & Interpolate → FFmpeg: processes video
- Video & Audio Edit → Upscale & Interpolate: optionally enhances
- Metadata & Export → MinIO: exports media