iArt.ai
AI 모션 그래픽 소프트웨어, 누구나 사용할 수 있는 After Effects 대안, 텍스트 프롬프트로 몇 분 안에 애니메이션·설명 영상 제작
복잡한 소프트웨어 없이 영상을 올리고 스타일을 선택하면 CutForge가 편집·애니메이션·음성·자막을 자동으로 만들어줍니다
Plus de logiciel compliqué. Charge ta vidéo parlée, choisis un style, CutForge forge le montage : coupes, illustrations animées, sons et sous-titres.
미디어Python · JavaScriptGitHub @gavoekoffi2온라인운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
개인 계정의 공개 저장소 소유자는 GitHub로 로그인해 운영자를 확인하고 소개를 직접 관리할 수 있습니다. 관리자 승인은 필요하지 않습니다.
로그인한 뒤 확인 파일을 만들어 이 저장소의 기본 브랜치에 30분 안에 커밋해 주세요. 다른 브랜치나 포크의 파일은 인정하지 않습니다. 파일을 올린 뒤 아래에서 확인하면 관리자 승인 없이 관리할 수 있습니다. 자격을 유지하려면 파일을 남겨 두세요.
로그인이 필요하면 로그인 화면으로 이동합니다. 로그인 후 이곳으로 돌아와 원하는 확인 버튼을 다시 눌러 주세요.
프로젝트 폴더에서 /nomorevibe verify를 실행하면 소유권을 확인하고 소개를 직접 관리할 수 있습니다.
1분 만에 확인하기logiciel de montage vidéo automatique
저장소 README에서 가져옴
README 전체를 불러오는 중입니다. 아래는 저장된 발췌입니다.
CutForge (ex-AutoEdit) - AI Video Editing SaaS Platform
AI-powered automatic video editing platform. Upload your talking-head video, choose a style, and CutForge forges the montage: smart cuts, illustrated motion design, AI B-roll, karaoke captions and pro sound design.
Backend:
• FastAPI (async Python web framework)
• PostgreSQL + SQLAlchemy (database)
• Celery + Redis (async job queue)
• OpenAI Whisper (AI transcription)
• auto-editor (silence removal)
• PySceneDetect (scene detection)
• MoviePy (video effects & compositing)README 전문 보기 ↗AI-powered automatic video editing platform. Upload your talking-head video, choose a style, and CutForge forges the montage: smart cuts, illustrated motion design, AI B-roll, karaoke captions and pro sound design.
AutoEdit/
├── backend/ # FastAPI + Celery backend
│ ├── app/
│ │ ├── api/v1/ # REST API endpoints
│ │ ├── models/ # SQLAlchemy models
│ │ ├── processing/ # Video processing pipeline
│ │ ├── workers/ # Celery async tasks
│ │ └── services/ # Auth, storage, payments
│ └── alembic/ # Database migrations
├── frontend/ # React + Vite + Tailwind
│ └── src/
│ ├── pages/ # Landing, Login, Dashboard, Editor, Pricing
│ ├── components/ # Reusable UI components
│ ├── api/ # API client
│ └── store/ # Zustand state management
├── nginx/ # Reverse proxy config
└── docker-compose.yml
Backend:
Frontend:
Infrastructure:
Deux pipelines coexistent — opt-in via pipeline_version sur le job:
V1 (stable, par défaut):
Upload → Whisper → auto-editor (silences) → PySceneDetect → MoviePy effects → Subtitles → Export
V2 (IA Afrique francophone — voir docs/VIDEO_PIPELINE_ARCHITECTURE.md):
Upload → Whisper (word-level) → Silence detect → EDL (filler words FR/EN)
→ BrollPlanner (Afrique) → OpenRouter image gen → Ken Burns FFmpeg
→ Overlays (intro/CTA) → FFmpeg final render (9:16 + captions + music)
| Mode | Description |
|---|---|
| TikTok viral | 9:16 + captions animées + B-roll IA + CTA |
| Business premium 🇸🇳🇨🇮🇹🇬 | Style africain moderne + musique sobre |
| Publicité locale | Restaurant / boutique / service local + CTA clair |
| Podcast propre | Suppression silences uniquement |
| Formation / éducatif | Captions lisibles + B-roll discret + horizontal |
| Collage Premium ✂️ | Métaphore visuelle + collage papier éditorial assemblé à l'écran (doc) |
| YouTube long 🎬 | Vidéos longues 16:9 : coupe « dernière prise », synchro image/son exacte, chapitres YouTube, sous-titres FR, cartes chapitre, SFX, −14 LUFS (doc) |
CutForge lit le transcript, décide où une illustration aide vraiment, quoi montrer et comment l'animer sur la voix — la vidéo originale reste le socle.
00:00 ───────── TALKING HEAD ─────────
00:35 ── PROCESSUS ── « produit → trafic → livraison »
00:41 ───────── TALKING HEAD ─────────
01:25 ── DIAGRAMME ── « client → serveur → entrepôt »
01:32 ───────── TALKING HEAD ─────────
offline est le défaut et ne contacte
rien ; les modes free et cloud sont optionnels et retombent toujours sur
l'analyse locale ;📖 docs/ILLUSTRATION_ENGINE.md ·
🖼️ exemples réels ·
⚖️ licences
Les modes V1 historiques (tiktok, youtube, podcast) restent supportés.
AUTOEDIT_AUDIT.md — audit technique complet, bugs corrigés, manques.docs/VIDEO_PIPELINE_ARCHITECTURE.md — pipeline V2, contrats, EDL, intégration HyperFrames / Remotion / video-use / OpenRouter.docs/COLLAGE_PREMIUM_BROLL.md — moteur Collage Premium B-roll (collage_assemble) : analyse sémantique, métaphore visuelle, verrou de style, contrôle qualité automatique, routage du type de B-roll.docs/ILLUSTRATION_ENGINE.md — moteur d'illustration : analyse du discours, score visuel, storyboard, 12 types de scènes, renderers et replis, SFX contextuels, synchronisation, modes offline/free/cloud.docs/ILLUSTRATION_ENGINE_EXAMPLES.md — sorties réelles du moteur, avec rendus.docs/ILLUSTRATION_ENGINE_AUDIT.md — audit de l'ancien moteur, contrats préservés, audit des licences.docs/THIRD_PARTY_LICENSES.md — licences tierces et décisions juridiques.docs/SFX_LICENSES.md — provenance des effets sonores (tous synthétisés).# Clone the repository
git clone https://github.com/gavoekoffi2/auto-edit.git
cd auto-edit
# Copy environment variables
cp .env.example .env
# Start all services
docker-compose up --build
Backend:
cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000
Frontend:
cd frontend
npm install
npm run dev
| Method | Endpoint | Auth | Description |
|---|---|---|---|
| POST | /api/v1/auth/signup | No | Create account |
| POST | /api/v1/auth/login | No | Login |
| POST | /api/v1/auth/refresh | No | Refresh token |
| GET | /api/v1/auth/me | Yes | Get profile |
| POST | /api/v1/videos/upload | Yes | Upload video |
| GET | /api/v1/videos | Yes | List videos |
| GET | /api/v1/videos/{id} | Yes | Get video |
| GET | /api/v1/videos/{id}/stream | Yes | Stream video |
| DELETE | /api/v1/videos/{id} | Yes | Delete video |
| POST | /api/v1/jobs | Yes | Create processing job |
| GET | /api/v1/jobs/modes | No | Catalogue des modes (v1 + v2) + defaults |
| GET | /api/v1/jobs/{id} | Yes | Get job status |
| POST | /api/v1/jobs/{id}/cancel | Yes | Cancel job |
| GET | /api/v1/jobs/{id}/download | Yes | Download result |
| POST | /api/v1/payments/checkout | Yes | Create checkout |
| POST | /api/v1/payments/webhook | No | FedaPay webhook |
| GET | /api/v1/payments/plans | No | Get pricing plans |
| Plan | Price (XOF) | Price (USD) | Features |
|---|---|---|---|
| Free | 0 | $0 | 2 videos/month, 5min max, 720p |
| Pro | 5,000 FCFA | $10 | Unlimited, 30min max, 1080p, all AI modes |
| Enterprise | 15,000 FCFA | $30 | Unlimited, no limit, 4K, API access |
Voir .env.example pour la liste complète. Essentielles :
DATABASE_URL, DATABASE_URL_SYNC — PostgreSQLREDIS_URL — Redis pour CelerySECRET_KEY — JWT signing key (obligatoire en prod, min 32 chars)FEDAPAY_SECRET_KEY, FEDAPAY_PUBLIC_KEY — paiementWHISPER_MODEL — taille du modèle (tiny/base/small/medium/large)Pipeline V2 (B-roll IA Afrique) :
PIPELINE_VERSION=v1|v2 — pipeline par défaut côté workerOPENROUTER_API_KEY — clé API OpenRouter (jamais en clair dans le repo)IMAGE_GENERATION_PROVIDER=openrouter / IMAGE_GENERATION_MODEL=google/gemini-2.5-flash-image-previewBROLL_STYLE=african_business_premiumVIDEO_RENDERER=ffmpeg|hyperframes|remotionENABLE_AI_BROLL, ENABLE_DYNAMIC_CAPTIONS, ENABLE_MUSIC, ENABLE_SFXcd backend
python -m pytest tests/ -q # 8 tests (EDL, broll planner, providers)
python scripts/smoke_pipeline_v2.py # smoke import + planner end-to-end
MIT