HonkTTS

HonkTTS is the text-to-speech service used by Unitystation to generate immersive in-game voices.
It is an optional module that runs alongside the game and is orchestrated by StationHub.

This repository contains:

• The Python TTS server (server/tts_server.py)
• Scripts to run and test the server
• A cross-platform .NET installer (installer/) that packages and sets up everything

Latest Installer

Download the latest installer build from GitHub Releases:

https://github.com/unitystation/honk_tts/releases

If you only want to use HonkTTS (not develop it), start there.

Contributing Prerequisites

Required:

• Git (clone/push workflow)
• .NET SDK 10 (project targets net10.0)

Useful depending on what you change:

• Python 3.10+ (for direct server work and local script testing)

Start Contributing

1. Clone the repo.
2. Build and run the local end-to-end flow:
   • Windows: dev.bat
   • Linux/macOS: ./dev.sh
3. This will:
   • Build the installer
   • Install into ./tts (including Python runtime and dependencies)
   • Run server tests (test_tts)

The dev flow is enough to start iterating even without a system Python install.
You can edit the checked-in scripts/source in this repo, rerun dev, and validate changes against the locally installed ./tts instance.

If you only want to build artifacts:

• Windows: build.bat
• Linux/macOS: ./build.sh

Project Layout

• server/ - TTS API server and runtime scripts
• installer/ - Native AOT .NET installer source
• build*.{bat,sh} - local build/package scripts
• dev*.{bat,sh} - quick contributor workflow scripts