#3535th of 51 in VoiceFor voice, speech and music
mlx-audio
Python library for TTS, STT and speech models on Apple Silicon via MLX
- Stars
- 8.1k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
MLX-Audio runs text-to-speech, speech-to-text, speech-to-speech, voice activity detection/diarization and music generation models on Apple Silicon using Apple's MLX framework. It offers a CLI, a Python API, an OpenAI-compatible REST API with a web interface, and a Swift package for iOS/macOS. Quantization from 3-bit to 8-bit is supported, and the README lists dozens of models such as Kokoro, Qwen3-TTS, Whisper and Parakeet.
Who it is for: Mac developers running local speech and audio models on Apple Silicon
Strengths
- Covers TTS, STT, diarization, enhancement and music generation in one library
- Quantized models (3-bit to 8-bit) for Apple Silicon, with many mlx-community weights
- CLI, Python API, OpenAI-compatible REST server and Swift package
- Supports voice cloning and streaming output on several models
Weaknesses
- Apple Silicon only, since it is built on MLX
- Some listed models are large (e.g. 35B MoE, 8B), so memory needs vary widely
- Server and web UI need the extra [server] install; no Docker provided
- README does not give RAM requirements or web port
What it needs
- GPU required
- Needs Apple MLX, Hugging Face model weights
- Models: Kokoro, Qwen3-TTS, Whisper, Parakeet, Chatterbox
Also in Voice
See all 51| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing and dictation with local speech models | 83 out of 100 |
| 2 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 80 out of 100 |
| 3 | whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription | 78 out of 100 |
| #4 | VoiceboxLocal voice studio for cloning, TTS, dictation and agent speech | 71 out of 100 |
| #5 | buzzOffline desktop app for transcribing and translating audio with Whisper | 71 out of 100 |