Vocalinux, an open source speech-to-text tool for Linux, released its 0.14 beta version on July 16. The update improves usability with customizable shortcuts and better support for Wayland sessions. It also expands options for remote transcription models.
The new release allows users to configure keyboard shortcuts through the Settings menu. Combinations now support Ctrl, Alt, Shift, and Super keys paired with letters or numbers.
On GNOME Wayland, text injection works again when using a bare XKB engine. The whisper.cpp engine also stops defaulting to all CPU cores on hybrid Intel and AMD laptops.
Remote API users can now access FunASR and SenseVoice models via OpenAI-compatible endpoints. The app remains fully local under the GPL-3.0 license, with voice data staying on the user's machine.
Vocalinux supports multiple engines including whisper.cpp as default, plus options for PyTorch, NVIDIA, VOSK, and remote servers. It runs from the system tray to dictate into any text field.
