Audio TTS
A 16MB Mac desktop app for text-to-speech and voice cloning, built almost entirely by AI ("vibe-coded") on top of Alibaba's Qwen3-TTS model.
🔗 Visit Audio TTS
Description
Most desktop text-to-speech apps are either bloated Electron wrappers or command-line tools that non-technical users won't touch. Audio TTS is a tiny (about 16MB) native-feeling Mac app that does text-to-speech, voice cloning, and even "voice design" — describing a voice in words and having the app generate it — while staying small and processing everything locally on your machine.
Built on Alibaba's Qwen3-TTS model and the Electrobun framework (a lightweight alternative to Electron), Audio TTS supports cloning a voice from an audio sample, designing new voices from text descriptions, using built-in "instruct" voices you can steer with instructions, and batch-generating multiple audio files at once. It auto-updates and keeps everything private and local rather than sending audio to a server. The project is notable for being described by its own author as "vibe-coded" — built largely through AI-assisted development — and has 157 GitHub stars and 46 commits; it's currently macOS-only, downloads its models automatically on first use, and its README doesn't specify a license.
💬 Our review
The short version: Audio TTS is an impressively compact, capable little Mac app — voice cloning AND voice design in 16MB is a genuine engineering feat — but the missing license and "vibe-coded" origin are worth knowing before you build anything serious on top of it.
Against KokoClone or VITS EVOlution, both cross-platform Python tools requiring a script or Gradio setup, Audio TTS's advantage is being a proper installable Mac app anyone can double-click — no Python environment to manage. Against a commercial app or API, it's free and fully local/private, at the cost of platform lock-in (Mac-only) and the uncertainty of an unspecified license, which matters if you plan to redistribute anything built with it. Its 157 stars suggest real interest despite being a young, single-purpose project. Worth trying if you're on a Mac and want a simple, private, no-subscription TTS/voice-cloning app; check the license situation directly with the author before using it in anything commercial.
💰 Pricing
📊 Global score
🤖 AI-enriched data
Gratuit, licence non précisée dans le README (à clarifier avant usage commercial)
Pros
App native Mac ultra-compacte (~16 Mo) malgré des fonctions avancées (clonage + design vocal)
Traitement 100% local et privé, aucun envoi audio vers un serveur
Génération par lot de plusieurs fichiers audio, mise à jour automatique
157 étoiles GitHub, intérêt réel pour un projet jeune
Cons
Exclusivement macOS, pas de version Windows/Linux
Licence non précisée dans le dépôt — statut légal à clarifier avant usage commercial
Projet jeune et "vibe-codé" (développement assisté par IA), stabilité à confirmer dans la durée