# wyoming-piper

> Self-hosted Wyoming protocol server for Piper TTS, providing neural text-to-speech for Home Assistant and Wyoming clients. Open-source, runs locally via Docker or Python.

- **Category:** Text to Speech
- **Licence:** MIT
- **Latest version:** v2.5.2
- **Open source:** oss
- **Languages:** Python
- **Platforms:** Docker
- **Tags:** Home Assistant, Docker, Text To Speech, Self Hosted, Voice cloning, Wyoming protocol, Neural tts, Voice synthesis, Omnivoice, Piper TTS

## Description

Wyoming Piper is a Wyoming protocol server for the Piper text-to-speech system. It enables self-hosted neural TTS with support for dozens of voices across many languages. The server speaks the Wyoming protocol, making it compatible with Home Assistant's Piper integration and other Wyoming clients. It can run via Docker, as a local Python installation, or as a Home Assistant add-on. An experimental OmniVoice backend adds voice cloning and design capabilities. A web UI allows managing custom voices. The project requires Python 3.10 or later and runs on Linux, with GPU acceleration available via a separate Docker image.

## Links

- [Homepage](https://github.com/rhasspy/wyoming-piper)
- [Docker](rhasspy/wyoming-piper)
- [GitHub Repository](https://github.com/OHF-Voice/wyoming-piper)
- [Docker Hub](https://hub.docker.com/r/rhasspy/wyoming-piper)

---
Source: https://selfhost.directory/project/wyoming-piper