oneword-ai 0.1.0

A minimalist one-word subtitle generator with Neobrutalism UI

🎬 OneWord AI - Subtitle Generator

Generate cinematic one-word subtitles from video/audio using Whisper AI

Perfect for creators making high-energy reels, shorts, and TikToks!

---

✨ Features

• 🎯 Three Subtitle Modes: One Word, Two Word Punch, Phrase Mode
• 🌍 Multi-Language: Auto-detect or specify (English, Hindi, Urdu, Spanish)
• 🤖 Multiple Models: Medium, Large, and Hindi2Hinglish (Oriserve/Prime) 🆕
• 📦 Python Package: Installable via pip with oneword-cli and oneword-web commands
• 💻 Local CLI: Robust command-line tool for batch processing
• 🌐 Web UI: Beautiful Neobrutalism-styled web interface
• ☁️ Cloud Ready: Works on Google Colab and Hugging Face Spaces
• 🐳 Docker Support: Containerized deployment

🚀 Quick Start

Installation

# Clone the repository
git clone https://github.com/Ambrishyadav-byte/OnewordAI.git
cd OnewordAI

# Install as a package
pip install -e .

Prerequisites: Ensure FFmpeg is installed on your system.

Usage

🖥️ CLI (Command Line)

See CLI.md for full documentation.

# Basic usage
oneword-cli -i video.mp4

# Full options
oneword-cli -i video.mp4 -m medium -lang hi -mode oneword

🌐 Web UI

# Start server & open browser
oneword-web

Features:

• Drag & drop file upload
• Real-time progress tracking
• Instant SRT download
• Responsive Neobrutalism design

☁️ Google Colab

1. Open OneWord_Colab.ipynb in Colab
2. Run all cells
3. Upload your video
4. Download your SRT!

🤗 Hugging Face Space

python app_gradio.py

Or deploy to Hugging Face Spaces for a hosted version!

📊 Subtitle Modes

Mode	Description	Use Case
One Word	Each word = separate subtitle	High-energy, attention-grabbing content
Two Word Punch	Groups of 2 words	Punchy, impactful messaging
Phrase Mode	Full sentence segments	Traditional subtitle style

🎨 Web UI Preview

The web interface features a stunning Neobrutalism design:

• Bold black borders
• Vibrant color palette
• Sharp shadows
• Grid background pattern
• Smooth animations

🐳 Docker Deployment

# Build image
docker build -t oneword-ai .

# Run container
docker run -p 8000:8000 oneword-ai

📁 Project Structure

minimalist-one-word-subtitle-generator/
├── onewordai/                    # Source code (package)
│   ├── core/
│   │   ├── __init__.py
│   │   └── engine.py          # Core subtitle generation logic
│   ├── api/
│   │   ├── __init__.py
│   │   └── main.py            # FastAPI backend
│   └── web/
│       ├── index.html         # Web UI
│       ├── style.css          # Neobrutalism styles
│       └── app.js             # Frontend logic
├── cli.py                     # CLI interface
├── app_gradio.py              # Gradio app for HF Spaces
├── OneWord_Colab.ipynb        # Google Colab notebook
├── Dockerfile                 # Docker configuration
├── requirements.txt           # Python dependencies
└── README.md

🛠️ Development

API Endpoints

• POST /upload - Upload video/audio file
• POST /process - Start transcription job
• GET /status/{job_id} - Check job progress
• GET /download/{job_id} - Download generated SRT

Requirements

• Python 3.8+
• FFmpeg
• PyTorch
• OpenAI Whisper
• FastAPI (for web server)
• Gradio (for HF Spaces)

💡 Tips for Creators

Video Editing Workflow

1. Generate SRT using OneWord AI
2. Import into your editor:
   • CapCut: Text → Local Captions → Upload
   • VN Editor: Text → SRT → Import
   • Premiere Pro: File → Import → Captions
3. Apply animations (Pop, Spring, Bounce)
4. Customize colors and fonts

Best Practices

• Use Tiny model for quick drafts
• Use Base model for production (best balance)
• Use Small model for technical/complex content
• One Word mode works best for 30-60 sec reels
• Enable language selection for multilingual content

🤝 Contributing

Contributions welcome! Open an issue or submit a PR.

Ideas for improvements:

• Auto-capitalization for emphasis
• Color-coded keywords
• Export with burned-in subtitles
• Batch processing multiple files

📜 License

MIT License - see license.txt

🤝 Credits

• OpenAI Whisper - Speech recognition
• FastAPI - Backend framework
• Gradio - ML web interfaces

Built with ❤️ by Ambrish

Follow for updates: @ambrish.yadav.1

---

⭐ Star this repo if you find it useful!