Velora – On-device macOS dictation (Whisper and a local LLM, no cloud)
Runs entirely offline on Apple Silicon, offering a private open-source Wispr Flow alternative.
Make Local AI Toys, Robots, Devices that work with a MacBook and an Arduino ESP32
Free local CharacterAI with voice cloning under 10s audio, plus ESP32 hardware integration.
Privacy-focused users, hardware hackers, ESP32 tinkerers
The voice pipeline currently supports MLX on any M1 through M5 chip. I used Whisper-Turbo for STT, Qwen3.5-9B-4bit for the LLM and Qwen3-TTS-0.6B-4bit for TTS.
The repo also has a Websocket Transport to add these voices to devices powered by the ESP32 via secure websockets.
Runs entirely offline on Apple Silicon, offering a private open-source Wispr Flow alternative.
MLX-powered local TTS plugin for OpenClaw—elegant but audience is Apple Silicon only.
Standardized MLX benchmarking when everyone's currently comparing engines manually.
Tauri GUI wrapper around mlx-lm—useful for Mac users, but local fine-tuning UIs already exist.
LiteRT beats MLX on Gemma memory while CoreML sips power on the Neural Engine.
Full MLX power in Ruby: lazy arrays, Metal GPU, transformer layers—but Ruby adoption risk.