AI-powered full-cast audiobooks — built for scale, accessibility, and open collaboration.
NeuralVox is an open project to generate audiobooks using speech-synthesis, with a focus on full-cast narration, automation, and accessibility. Inspired by LibriVox, we aim to use modern AI models to produce high-quality readings of public domain texts - and eventually, user-submitted works.
We're currently building:
- End-to-end pipeline for text-to-audiobook conversion
- Voice assignment system to map characters to distinct AI voices
- Modular synthesis components for speech generation, pacing, and emotion control
- Public audiobook library for classic literature
- Interfaces for authors and contributors
This project is in active development and is not yet ready for public deployment. If you're interested in contributing to early infrastructure, reach out or watch this org for repo announcements.
- Join the waitlist for early access
- Discuss on the Discord server
- Follow @neuralvox for updates
- Star or watch repos to track releases
- Contributions welcome once repos are public