🚀 I built an offline AI speech-to-text app that runs entirely on your GPU. Most voice-to-text to...🚀 I built an offline AI speech-to-text app that runs entirely on your GPU. Most voice-to-text to...
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Alessandro's avatar
Alessandro Federici
pro
• Jan 24
🚀 I built an offline AI speech-to-text app that runs entirely on your GPU.
Most voice-to-text tools today rely on cloud APIs, subscriptions, and constant internet access.
AF Speech is a Desktop app that converts voice into text using local AI, powered by the user’s own NVIDIA GPU.
No cloud. No per-minute costs. No internet required.
🧠 The AI model runs offline using Whisper Large-v3, accelerated with CUDA, delivering fast and accurate transcription directly on the PC.
To make it usable in real-world scenarios, I also implemented:
- global hotkeys for instant dictation
- system tray integration
- GPU VRAM protection for gaming/workstation setups
- visual listening overlay
🔐 All audio data stays on the local machine — 100% privacy.
Check it out
AI Voice DevelopmentCUDAGoogle Antigravity
Back to feed
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started