TBPN-Transcripts
TBPN Transcripts
Transcripts and summaries generated from TBPN episodes through OpenAI Whisper, Llama 3.1, and LanguageTool.
Key Features • Installation • Usage • Documentation • Support • Contributing • Credits • License
Key Features
- Downloads every published episode and converts it to audio.
- Ability to download all pre-created transcripts.
- Create transcripts for every episode using OpenAI’s Whisper.
- Generate a summary from the transcript using Ollama and Llama.
- Correct spelling and grammatical errors using LanguageTool.
Installation
Requires Python, Ollama, and ffmpeg.
git clone https://github.com/willtheorangeguy/TBPN-Transcripts.git
cd TBPN-Transcripts
pip install -r requirements.txt
Full prerequisites, including GPU-accelerated Whisper, are in docs/usage.md.
Usage
python main.py <year>
That runs the whole pipeline end to end. Episodes live under 2025/.
Documentation
Full documentation lives in docs/:
Quickstart · Installation · Configuration · Architecture · Pipeline · FAQ · Troubleshooting
Support
Open a GitHub Discussion or file an issue.
Contributing
Please contribute using GitHub Flow. Create a branch, add commits, and open a pull request.
Please read CONTRIBUTING for details on our CODE OF CONDUCT, and the process for submitting pull requests to us.
Changelog
See the CHANGELOG file for details.
Contributions welcome. See the org-wide Contributing Guide and Code of Conduct.
Credits
 |
 |
 |
 |
 |
|---|---|---|---|---|
| Python | OpenAI Whisper | Ollama | LanguageTool | TBPN |
| Web - Donate | GitHub | GitHub | GitHub | Web |
License
The pipeline code is MIT — see LICENSE.md.
The transcripts are not. They are machine-generated from recordings of TBPN episodes, and the words belong to their speakers and rights holders. See CONTENT_LICENSE.md before reusing any of it.