tl;dl is a podcast summarization tool for Google Podcasts. (probably buggy, since this was a side project just for myself)
It uses;
- streamlit for user interface
- whisper for speech-to-text
- ollama for summarization
- TTS for text-to-speech
ollama must be installed and served
ollama run llama3:instructgit clone https://github.com/makermotion/tldl.git
cd tldl
pip install -r requirements.txtnote: python version 3.10.12
To use the podcast summarizer, you need to run the UI via: (you need to be on the project directory)
streamlit run app.py- tl;dl heavily relies on the web scraping, links for podcast soundfile links are collected from the Google Podcasts website from some specific html tags. So all podcasts may not be supported. If you want to contribute you can file an issue for not supported podcasts or you can contribute to the codebase.
- Also there may be a script for tl;dl to be used as a command line tool.
- dockerize the application
This script is released under the MIT License. See the LICENSE file in the repository for full details.