Press a key, speak, and polished text lands in any app. 100% local speech-to-text, optional AI polish, native GTK 4. The community Linux port of FluidVoice.
Hey Product Hunt!
SayItErmano is 100% local voice dictation for Linux. Press a key, speak, and your words are typed straight into whatever app you're in — editor, browser, terminal, or chat. Speech-to-text runs entirely on your machine with Whisper, plus an optional AI polish step to tidy up the result.
What I focused on:
- Privacy first: no telemetry, no network listeners by default, API keys only via environment variables
- Works everywhere via keystroke insertion, on both X11 and Wayland (NVIDIA CUDA or CPU)
- Native GTK 4 History and Settings windows with a live waveform + streaming transcription overlay
- 108 spoken-punctuation aliases
Full disclosure: it's an unofficial community Linux port of the excellent macOS app FluidVoice, licensed under GPLv3. Feedback and contributions very welcome!