Developer Tools●●Solid
Local TTS for OpenClaw on Apple Silicon (MLX-Powered, Zero Setup)
MLX-powered local TTS plugin for OpenClaw—elegant but audience is Apple Silicon only.
Niche GemCozy
ZacharyZZ
203mo ago
Back to browse
GitHub Repository
Make Local AI Toys, Robots, Devices that with a MacBook and an Arduino ESP32
121 starsTypeScript
I built a free CharacterAI that runs locally
by akadeb·Mar 25, 2026·8 points·4 comments
AI Analysis
●●●BangerZero to OneWizardrySolve My Problem
Free local CharacterAI with voice cloning under 10s audio, plus ESP32 hardware integration.
Strengths
- •Fully local pipeline: Whisper-Turbo + Qwen3.5-9B + Qwen3-TTS-0.6B on Apple Silicon
- •ESP32 websocket transport enables physical toys and robots, not just software
- •Clone any voice with under 10 seconds of audio, supports 10+ languages
Weaknesses
- •Apple Silicon only via MLX, no Windows or Linux support yet
- •34 stars suggests early stage, may have rough edges in production use
Category
Target Audience
Privacy-focused users, hardware hackers, ESP32 tinkerers
Post Description
CharacterAI has over 20M active users and lets people talk to AI voice clones. I built a local alternative that clones any voice with <10s of audio and can talk fluently in over 10 languages, and runs locally on your Macbook or Mac mini.
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.
Similar Projects
Developer Tools●●Solid
Courtyard – Open-source macOS app for local MLX fine-tuning Text
Tauri GUI wrapper around mlx-lm—useful for Mac users, but local fine-tuning UIs already exist.
Solve My ProblemShip It
tuwenbo0120
103mo ago
AI/ML●●Solid
MLX-Ruby – Ruby Bindings for Apple's MLX ML Framework
Full MLX power in Ruby: lazy arrays, Metal GPU, transformer layers—but Ruby adoption risk.
WizardryDark HorseNiche Gem
skryl
113mo ago
Developer Tools●●●Banger
Fleet Watch – preflight guard for local AI inference on Apple Silicon
GPU working set estimation catches memory overcommit before your 7B model swaps to SSD.
Solve My ProblemBig BrainNiche Gem
cjarchivist
211mo ago
AI/ML●●●Banger
SwiftLM – Qwen Chat on iPhone, 100B+ Moe on M5 Pro 64GB (Native Swift)
Native Swift inference with SSD streaming runs 100B MoE models without kernel panics.
WizardryNiche Gem
aegis_camera
122mo ago
AI/ML●●Solid
PrivateWhisper – Run Whisper locally on macOS (offline transcription)
Packages Whisper into a one-click Mac app that records system audio, batch-processes files, and exports SRT/MD/DOCX — all kept on-device. Very useful if you hate terminal hacks and want privacy-first transcription, but it’s not novel (several local Whisper GUIs already exist) and the Apple‑Silicon‑only restriction plus limited detail on model sizes and performance hold it back.
Solve My ProblemNiche GemSlick
matyashajek
213mo ago