We built an AI podcast generator that turns any content—PDFs, URLs, or prompts—into realistic, multi-speaker podcasts using cloned voices. Unlike NotebookLM, users can choose or even record their own voices for hosts and guests, creating highly personalized and engaging audio content. Powered by Qwen for script generation and Higgs 2.5 for voice synthesis, our system makes content creation faster, more expressive, and more human.
Built With
- fastapi
- javascript
- python
- uvicorn
Log in or sign up for Devpost to join the conversation.