Vosk CLI Dictation

Real-time voice dictation that revolutionizes your Linux workflow

Transform your voice into text instantly in any application. Built with Python and Vosk, completely open-source and privacy-respecting.

Get Started Now View on GitHub
v0.0.4-alpha Current Version
2 Languages
100% Offline & Private

See it in action

Experience the power of voice dictation directly in your terminal

Real-time
High accuracy
Private by design

Key Features

Real-time transcription

Your voice becomes text instantly in any active application on your system.

Multi-language support

Works out-of-the-box with French and English, with an easily extensible configuration.

Total control

Voice commands, global keyboard shortcuts, and manual commands for a tailored experience.

Advanced configuration

Customize everything via config.yaml: theme, shortcuts, recognition aliases and much more.

Complete punctuation

Dictate all punctuation marks with natural voice commands in each language.

100% offline

No data leaves your machine. Based on Vosk, an open-source recognition engine.

Total Configuration

The config.yaml file gives you unprecedented control over your experience

Snippet of the config.yaml file showing hotkeys and theme color settings.

Theme and shortcuts

Fully customize the interface with your favorite colors and define your own keyboard shortcuts for each action.

  • Customizable colors for each message type
  • Configurable global keyboard shortcuts
  • Personalize start & stop voice commands
Snippet of the config.yaml file showing English language settings.

Advanced settings

Fine-tune voice recognition with custom aliases, adapted voice commands, and optimized audio parameters.

  • Alias system to correct frequent errors
  • Customizable voice commands per language
  • Extensible punctuation dictionary

Adaptive Interface

A native experience in your language, with an interface that adapts automatically.

French interface

Screenshot of the French command-line interface.

English interface

Screenshot of the English command-line interface.
Language selection at startup
Translated help messages and status
Native voice commands per language

Getting Started

Set up your voice dictation environment in minutes on Debian/Ubuntu.

1

System Dependencies

Install the necessary dependencies for audio and keyboard control:

sudo apt-get update && sudo apt-get install python3-pip python3-venv portaudio19-dev gettext xdotool pulseaudio-utils
2

Clone and Setup

Clone the repository and create a virtual environment:

git clone https://github.com/RonanDavalan/vosk-cli-dictation.git
cd vosk-cli-dictation
python3 -m venv venv
source venv/bin/activate
3

Install Dependencies

Install the required Python packages:

pip install -r requirements.txt

Download your language models from the Vosk models page and place them in the vosk-model/ directory.

4

Launch

Start dictating with your voice:

python3 src/main.py -l fr

Pro tip: Add a shell function to your ~/.bashrc for quick access from anywhere!

Ready to transform your workflow?

Start Your Journey