Teletext IA

This App is hosted on Streamlit Community Cloud and is accessible at the following address:

• teletext.streamlit.app

Description

Teletext IA is an experimental project for improving video/audio accessibility with AI. It combines research notebooks (transcription, subtitles, descriptive captions) and a Streamlit prototype dashboard.

Table of Contents

• Teletext IA
  • Description
  • Table of Contents
  • 🎯 Project objective
  • 👥 Target audience
  • ⚙️ Current project content
  • 🗂️ Repository structure
  • 🚀 Quick start
  • 🐳 Installation & deployment
  • 🥽 Security
  • 📰 Changelog
  • 🩷 Acknowledgements
    • Environment
  • 🧪 Project Status
  • 🔒 License
  • 🤝 Contributing
  • 👤 Author

🎯 Project objective

Build practical tools to generate accessibility outputs from media files:

• subtitles (SRT)
• text transcription
• scene-level descriptive captions
• multimodal combinations (audio + visual context)

👥 Target audience

• Developers working on media accessibility tools
• Students or practitioners exploring Whisper and vision-language pipelines
• Teams prototyping inclusive video/audio workflows

⚙️ Current project content

• dashboard/: Streamlit prototype interface
  • upload audio/video
  • select one or more target outputs
  • current pipeline is a placeholder (simulated processing + ZIP output)
• notebooks/: research and experimentation notebooks
  • Whisper-based transcription and subtitle generation
  • folder transcription workflow
  • descriptive caption generation with scene detection + image captioning model
• project governance files: contribution, security, license, changelog, acknowledgements

🗂️ Repository structure

teletext-ia/
├── dashboard/
│   ├── src/
│   ├── requirements.txt
│   └── streamlit_app.py
├── docs/
│   └── images/
├── notebooks/
│   ├── drafts/
│   ├── src/
│   ├── descriptive_captions.ipynb
│   ├── folder_transcription.ipynb
│   ├── subtitle.ipynb
│   └── transcription.ipynb
├── .gitignore
├── ACKNOWLEDGEMENTS.md
├── CHANGELOG.md
├── CODE_OF_CONDUCT.md
├── CONTRIBUTING.md
├── LICENSE.md
├── README.md
└── SECURITY.md

🚀 Quick start

1. Clone the repository.
2. Install dashboard dependencies.
3. Run the Streamlit app.
4. Open notebooks for experimentation and iteration on AI pipelines.

🐳 Installation & deployment

The project is planned to be deployed:

• Via Docker for simple and reproducible containerisation

$ docker build -t streamlit-dashboard .
$ docker run -p 8501:8501 streamlit-dashboard

• Directly on the server by running the Streamlit application

$ streamlit run streamlit_app.py 

🥽 Security

• See SECURITY.md for vulnerability reporting guidelines.

📰 Changelog

Track all notable project changes in CHANGELOG.md.

Recommended:

• Follow a consistent format such as Keep a Changelog
• Create an entry for each release
• Include Added, Changed, Fixed, and Removed sections when relevant

🩷 Acknowledgements

• See ACKNOWLEDGEMENTS.md for people, tools, libraries, and communities that support this project.

Environment

• Python ≥ 3.13.5
• Dependencies listed in requirements.txt
• Additional notebook dependencies installed according to experiment needs

🧪 Project Status

• 🔬 Status: experimental prototype
• ✅ Working today: Streamlit upload flow and ZIP export, notebook experiments
• 🚧 In progress: implementation of real production pipeline in dashboard/src/

🔒 License

• See LICENSE.md.

🤝 Contributing

Contributions are welcome.

• See CONTRIBUTING.md
• Code of conduct available in CODE_OF_CONDUCT.md.

👤 Author

Gauthier Rammault