RawriisSTT

<p align="center"> <img src="assets/RawriisIcon.png" width="128" alt="RawriisSTT icon"> </p> <p align="center"> Speech-to-Text for VRChat - local, private, and controller-friendly </p> <p align="center"> <a href="https://github.com/hiccup444/RawriisSST/releases"><img alt="Downloads" src="https://img.shields.io/github/downloads/hiccup444/RawriisSST/total?label=Downloads"></a> <a href="https://github.com/hiccup444/RawriisSST/releases/latest"><img alt="Latest release" src="https://img.shields.io/github/v/release/hiccup444/RawriisSST?label=Release"></a> <img alt="Platform" src="https://img.shields.io/badge/Platform-Windows%20%7C%20Linux-blue"> <a href="https://discord.gg/4bG2dEfRSJ"><img alt="Discord" src="https://img.shields.io/badge/Discord-Join-5865F2?logo=discord&logoColor=white"></a> </p>

---

RawriisSTT listens to your microphone and sends your speech to the VRChat chatbox over OSC — no subscription required. Designed for minimal performance impact, it uses efficient engines like faster-whisper for quick, local transcription. Choose from multiple local or cloud speech engines, bind push-to-talk to a keyboard key or SteamVR controller button, and optionally have messages read back through the text-to-speech voice of your choice.

---

https://github.com/user-attachments/assets/deaae2fa-ed47-4ca5-b076-ed832f73e4df

Documentation

| Guide | Description | |---|---| | Windows Setup | Download, install, and get started on Windows | | Linux Setup | Running from source on Linux | | STT & TTS Engines | Per-engine setup for Whisper, Azure, Vosk, ElevenLabs, Polly | | Custom ElevenLabs Voices | Creating and cloning voices on ElevenLabs to use in the app | | SteamVR Bindings | Binding controller buttons to push-to-talk | | Building from Source | Compiling your own exe |

---

Features

🎙️ Multiple STT engines

• Whisper (local, CPU or CUDA) - no internet, no API key
• Azure Speech (cloud) - fast and accurate with an Azure subscription
• Vosk (offline) - lightweight, no internet required
• System STT - uses your OS speech recognition

🔊 Text-to-Speech playback

• System TTS (pyttsx3 / espeak) - zero setup
• ElevenLabs - high-quality AI voices with full parameter control
• Amazon Polly - neural voices via AWS
• Configurable delay before playback begins
• Linux: one-click virtual audio cable creation (PipeWire/PulseAudio)

💬 VRChat chatbox integration

• Sends transcriptions via OSC to /chatbox/input
• Broadcasts a listening state to your avatar parameters (stt_listening)

🎮 SteamVR controller bindings

• Bind push-to-talk to any controller button through the SteamVR input system
• Works without a headset connected (background app mode)
• Default bindings for Meta Quest, Valve Index, HTC Vive, WMR, HP Reverb G2, and generic controllers

⌨️ Hotkeys

• Keyboard push-to-talk (hold or toggle mode)
• PTT push/release sound effects
• Global hotkeys to stop or re-send the last TTS message

🧠 Voice Activity Detection

• WebRTC VAD - automatic speech detection with configurable aggressiveness and silence timeout

📋 TTS queue & smart split

• Queue messages and play them sequentially
• Automatically split long messages at word boundaries (respects VRChat's 144-character chatbox limit)

💾 Presets

• Save and load full configuration snapshots to switch between setups instantly

📦 In-app model downloads

• Download Whisper and Vosk models directly from the UI - no manual file management

🎨 Themed UI built with PyQt6 - dark mode or pastel pink light mode, switchable live from Settings

---

Quick Start (Windows)

1. Download the latest RawriisSTT-vX.X.X.zip from Releases.
2. Extract and run RawriisSTT.exe.
3. Enable VRChat OSC: Options → OSC → Enable.
4. In Settings → Speech-to-Text, pick a Whisper model and download it.
5. Back on the main window, select your microphone and click Launch Whisper.
6. Click Start Recording - your speech will now appear in the VRChat chatbox.

See Windows Setup for a full walkthrough.

---

Credits

• faster-whisper - efficient Whisper inference via CTranslate2
• pyopenvr - Python bindings for OpenVR / SteamVR
• python-osc - OSC messaging
• webrtcvad - Voice Activity Detection
• Vosk - offline speech recognition
• PyQt6 - UI framework