I'm excited to announce the first release of Local-NotebookLM, a powerful AI-driven tool that transforms PDF documents into engaging podcasts using local LLMs and TTS models! 🚀
What's included in this release: ✨
- **Complete PDF processing pipeline** 📄 - extract text from PDFs and convert them into conversational content
- **Flexible model support** 🤖 - works with OpenAI, Groq, Azure, LMStudio, Ollama, and custom endpoints
- **Multiple output formats** 🎙️ - generate summaries, podcasts, articles, or interviews
- **Customizable styles and lengths** 📏 - from short, casual summaries to in-depth technical discussions
- **Text-to-Speech integration** 🔊 - convert transcripts to natural-sounding audio with multiple voices
- **Command-line interface** 💻 - easy to use with comprehensive options
- **Programmatic API** 🧩 - integrate into your own Python projects
- **FastAPI server** 🌐 - access functionality through a web API
Getting started: 🏁
bash
pip install Local-NotebookLM
python -m local_notebooklm.start --pdf my_document.pdf
Why use Local-NotebookLM? 🤔
This tool bridges the gap between static PDFs and dynamic audio content, making information more accessible and engaging. Perfect for researchers, content creators, educators, or anyone who wants to consume PDF content in a more accessible format.
What's next: 🔮
I'm planning future releases to improve audio quality, add more customization options, and expand model compatibility. Your feedback is essential for guiding my development priorities!
---
Check out the example podcast in the repository to hear what Local-NotebookLM can produce. For detailed documentation, configuration options, and usage examples, please refer to the README.md.
Best, Gökdeniz Gülmez 🙏