Features
100% offline voice dictation (no cloud dependency)
Multiple engine support: whisper.cpp (default), Whisper, VOSK, Remote API
GPU acceleration via Vulkan (AMD, Intel, NVIDIA)
Works on X11 and Wayland display servers
Toggle mode (double-tap) and push-to-talk activation
Real-time transcription with low latency
Audio feedback sounds for recording start/stop
Remote API engine for OpenAI-compatible or whisper.cpp server
Silero Voice Activity Detection (neural VAD)
Configurable model size, language, and parameters
Guided whisper.cpp model selection (v0.13.0-beta)
Keyboard hotplug support (v0.13.0-beta)
IBus engine hardening and keyboard layout preservation
One-command interactive installer (Ubuntu, Fedora, Debian, Arch, openSUSE)
Voice dictation in every desktop and mobile app
AI auto-edits: removes filler words and fixes grammar
4x faster than typing (up to 220 wpm)
Cross-platform: Mac, Windows, iPhone, Android
Command mode for hands-free text editing
Custom voice commands and dictionary
Support for 100+ languages
Privacy mode (zero data retention)
HIPAA-ready on all plans (enforced on Enterprise)
SOC 2 Type II and ISO 27001 compliance (Enterprise)
Team collaboration and shared dictionaries
Student discount (3 months free, 50% off Pro)
Automatic microphone selection
Web demo available via browser
