Roadmap
Where QuillVoice is headed. Everything here keeps the same promise: local-first, no subscription, your voice never leaves your PC.
Where QuillVoice is headed. Everything here keeps the same promise: local-first, no subscription, your voice never leaves your PC.
On-device speech recognition with whisper.cpp. Audio never leaves your machine.
Normal, compact, and mic-only widget that types into any focused app without stealing focus.
Say "comma", "new line", "open paren" — spoken punctuation and formatting commands.
Teach Quill your names and jargon, and auto-fix common mishears.
CUDA acceleration for NVIDIA GPUs with automatic CPU fallback.
Multilingual models transcribe dozens of languages, fully offline.
Licenses are signed and verified on-device — no activation server, works offline.
Quill switches dictation profiles automatically based on the app you're in.
Quill learns from how you edit your transcripts and adapts over time.
Teach Quill your world — a richer way to add names, terms, and pronunciations.
Optional on-device LLM that tidies grammar and filler — fully on-device, nothing sent to the cloud.
A native macOS build of QuillVoice.
Local stats on words dictated, time saved, and usage trends.
A signed "no data leaves the device" attestation for security reviews and procurement.
Dictate on the go with a companion app for iPhone.
Central management of seats and settings for teams.
Share custom vocabulary and replacements across a team.