Voice-controlled automation with STT/TTS and AI vision.
| File | Description |
|---|---|
| voice_keyboard.py | Type with voice commands |
| voice_mouse.py | Control mouse with voice + AI |
| voice_assistant.sh | Interactive voice assistant |
| dictation_mode.py | Continuous voice dictation |
# (optional but recommended) Auto-configure voice in .env
streamware --setup --mode balance
# Or only TTS settings:
streamware --setup tts
# Voice keyboard
sq voice-keyboard type "wpisz hello world"
# Voice mouse (AI finds buttons!)
sq voice-click "kliknij w button OK"
# Text to speech
sq voice speak "Hello, I am Streamware"
# Listen for voice
sq voice listen
# System tools
sudo apt-get install xdotool espeak scrot
# Python packages (auto-installed)
pip install SpeechRecognition PyAudio pyttsx3
You can fine-tune voice behaviour via .env:
SQ_STT_PROVIDER – STT engine (google, whisper_local, whisper_api)SQ_WHISPER_MODEL – Whisper model size (tiny, base, small, medium, large)SQ_TTS_ENGINE – TTS backend (auto, pyttsx3, espeak, say, powershell)SQ_TTS_VOICE – preferred voice name (substring, e.g. polski)SQ_TTS_RATE – speech rate (words per minute)| Command | Action |
|---|---|
| “wpisz hello” / “type hello” | Types “hello” |
| “naciśnij enter” / “press enter” | Presses Enter |
| “kliknij w button” / “click button” | AI finds and clicks |