---
title: README
emoji: 🎭
colorFrom: gray
colorTo: yellow
sdk: static
pinned: false
---

**Artificial Humanity** builds deeply expressive, local-first speech synthesis — the framework
layer between text and genuinely dramatic spoken performance, running entirely on-device.

## 🎭 Prosodia — the on-device dramatic audiobook engine

Most text-to-speech reads. Prosodia *performs*. Every book is staged as a production:
a **Director** (an on-device LLM) reads ahead and annotates each passage with emotional
direction — valence, arousal, tension; an **Actor** (neural TTS) performs those notes through
a Rust synthesis core; a **Stage** coordinator keeps the audio gap-free. No cloud, no
telemetry — no page of your book leaves your hands.

## 🎙️ Sonora — training the voices

Our voice-actor training project. The first trained voice is live in the
[Sonora model registry](https://huggingface.co/artificial-humanity/Sonora) —
fidelity-verified ONNX and LiteRT/TFLite artifacts, including a mobile-ready split-graph
export. You can hear it in the [Prosodia Space](https://huggingface.co/spaces/artificial-humanity/Prosodia).
Emotional directability (VAT conditioning) and a multi-voice casting grid are in active
development.

---

🔗 [GitHub](https://github.com/Artificial-Humanity) · 🎧 [Hear the voice](https://huggingface.co/spaces/artificial-humanity/Prosodia) · 🎙️ [Sonora models](https://huggingface.co/artificial-humanity/Sonora)