Skip to content

Adams-404/Oralo

Repository files navigation

🎙️ Oralo (Voice Typer Tool)

Oralo is a global hotkey voice-to-text tool designed for Linux. Hold down a keyboard button, speak into your mic, and let the backend transcribe and clean up your audio with AI before automatically typing the result into whichever app was focused on your screen!

Powered by FastAPI, Groq Whisper (for ultra-fast transcription), and Groq LLaMA 3.1 (for contextually fixing filler words and fixing grammar).


🗺️ How it Works

  1. Press & Hold Your Hotkey (Default: Right Ctrl) ➔ Script records audio with arecord.
  2. Speak into your microphone.
  3. Release Hotkey ➔ Script sends audio to the FastAPI Backend.
  4. Backend Process:
    • Sends audio to Groq Whisper to structure text.
    • Feeds text to LLaMA to remove "ums", "uhs", and fix punctuation.
  5. Auto-Type ➔ Returns clean text to the client, which copies to your clipboard with wl-copy and simulates Ctrl+V pasting using evdev for full GNOME Wayland compatibility!

🚀 Quick Desktop Installation (Recommended)

Download the latest pre-compiled bundle to use Oralo instantly as a native app on your system framing setups:

  1. Download Oralo_Linux_x64.tar.gz from the Latest Releases tab.
  2. Extract the archive to any folder on your desktop context layouts frame accurate triggers transparent trigger context bound setups trigger context setups.
  3. Open a terminal inside the extracted folder and run the installer loader:
    ./install.sh
  4. Launch: Search for Oralo inside your Applications sidebar search setups transparently grids framing trigger structures setups accurately triggers systems correctly.

🛠️ Run From Source (Developers)

If you prefer building and managing scripts manually within virtual environments:

🟢 Step 1: Install source launcher shortcut

./install.sh

🟡 Step 2: Open and Configure

  1. Open Oralo in your system applet configurations dashboard framing securely framing triggers environments configs.
  2. Safe configurations settings and click ▶️ Start Oralo layouts setups framing.

To run setups manually with a single view:

🔑 Step 1: Set Your API Key

echo "GROQ_API_KEY=your_api_key_here" > .env

🟢 Step 2: Start the System

./start.sh

🔧 Troubleshooting

🛑 "No input devices found" on Wayland

If the script can't read devices, add yourself to the input group so you don't need a wrapper:

sudo usermod -aG input $USER

(Requires a full Desktop Log out & back in to apply values permanently)

💡 Enable Automatic Ctrl+V Simulation

To let the script automatically press Ctrl+V on Wayland automatically, grant your input group permissions to create virtual devices permanently:

echo 'KERNEL=="uinput", GROUP="input", MODE="0660"' | sudo tee /etc/udev/rules.d/99-uinput.rules
sudo udevadm trigger

🛑 Recording is empty or fails

Test your microphone manually to verify it records:

arecord -f S16_LE -r 16000 -c 1 -d 3 test.wav && aplay test.wav

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors