vox

A universal AI toolkit for high-performance Speech-to-Text (STT) and Text-to-Speech (TTS) processing, designed for low-latency and easy model integration.

What is vox?

A universal AI toolkit for high-performance Speech-to-Text (STT) and Text-to-Speech (TTS) processing, designed for low-latency and easy model integration.

Overview of vox

What it is

vox is a universal AI toolkit for high-performance Speech-to-Text (STT) and Text-to-Speech (TTS) processing. It is a single Rust binary that runs locally, without Python or cloud services. It provides multiple TTS backends, Whisper-based STT, an MCP server with 14 tools, and a command to configure AI tools on the machine.

Key features

  • Multiple TTS backends: pocket-tts (candle), piper (ONNX), qwen-native (candle), macOS say command, and kokoro (ONNX, optional build).
  • Speech-to-Text via Whisper (candle) supporting 99 languages, with CPU, Metal, or CUDA acceleration.
  • Voice cloning support in pocket-tts (8 predefined voices) and qwen-native (requires a reference recording).
  • A daemon mode that keeps models loaded between calls, with a configurable idle timeout (default 300 seconds).
  • An MCP server with 14 tools; the server does not use the daemon and keeps models loaded in its own process.
  • Environment variable VOX_TIMINGS=1 prints timing of each step on stderr.
  • Pre-built binaries for macOS (Apple Silicon), Linux (x86_64, ARM64), and Windows (CPU build).

Use cases

  • Local voice for AI coding agents: text-to-speech and speech-to-text without cloud dependencies.
  • Low-latency TTS for applications requiring fast synthesis, such as voice assistants or accessibility tools.
  • Offline STT for transcribing audio in 99 languages, suitable for privacy-sensitive environments.
  • Voice cloning for personalized speech synthesis using reference recordings.

Getting started

bash# Quick install (macOS Apple Silicon, Linux x86_64 and ARM64, WSL2)curl -fsSL https://raw.githubusercontent.com/rtk-ai/vox/main/install.sh | sh# Custom install dir (no sudo needed)curl -fsSL https://raw.githubusercontent.com/rtk-ai/vox/main/install.sh | VOX_INSTALL_DIR=~/.local/bin sh# Homebrew (macOS Apple Silicon, Linux)brew install rtk-ai/tap/vox

License and pricing

Licensed under Apache-2.0. The project is open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/rtk-aivox.svg)](https://yunodb.com/tools/rtk-aivox?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/rtk-aivox?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/rtk-aivox.svg" alt="Featured on YunoDB" height="20"></a>