Linux Dictation

Simple speech-to-text dictation tool hacked together for Linux machine based on RealtimeSTT library.

Features

• Real-time speech recognition using Whisper models through RealtimeSTT
• Automatic text pasting to the currently focused input field
• Automatic shutdown on inactivity

Files

• linux_dictation.py: Main script that handles speech recognition and text processing
• gui_printing.py: Utility for pasting text to the currently focused input field
• simple_stt_demo.py: Simple demonstration of the speech-to-text capabilities

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/linux-dictation.git
   cd linux-dictation

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate

3. Install the required packages (check your torch versions before executing):

   pip install torch==2.5.1+cu121 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu121
   pip install -r requirements.txt

Usage

Run the dictation tool with python linux_dictation.py

Wait for the "speak now" prompt, then start speaking. If you execute the file for the first time, the code will download the necessary Speech-to-Text models from Silero VAD. Once downloaded, the tool is ready to transcribe your speech and paste it into the currently focused input field.

Terminal Alias

Add this alias to your .bashrc for quick access:

alias start_dictation="cd ~/path/to/SpeechToText && source ~/path/to/virtual_environment && python linux_dictation.py"

Replace /path/to/dir/ with your paths.