themanyone/voice_typing

State-of-the-art offline (or networked) voice typing everywhere + text terminals (Linux or WFL session on Windows.) with a simple bash script. Usable with X. Does not require X.

46
/ 100
Emerging

Implements a modular client-server architecture where `voice_typing` runs Whisper locally (loading/unloading per utterance for minimal memory), while `voice_client` optionally connects to a persistent `whisper-server` instance for faster continuous dictation—supporting GPU acceleration via CUDA and quantized models down to 48 MiB VRAM. Text injection uses `ydotool` virtual keyboard daemon, enabling hands-free input across any window manager or headless environment, with voice activity detection via `sox` and optional Silero VAD for improved accuracy in noisy conditions.

154 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

154

Forks

9

Language

Shell

License

GPL-2.0

Last pushed

Mar 03, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/themanyone/voice_typing"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.