Installation

Mockingbird is a PySide6 desktop app for Windows 10/11 and Linux (x86_64). Download the build for your OS from thereleases page.

Requirements

• Windows 10/11 or Linux (x86_64).
• NVIDIA GPU with CUDA — to accelerate speech recognition (optional, but recommended).
• ~2–4 GB of disk space for the faster-whisper large-v3-turbo model.

Windows

1. Download Mockingbird-setup.exe from the latest release.
2. Run the installer and follow the wizard steps.
3. Launch the app from the Start menu.

If CUDA is not detected, the app automatically falls back to CPU — recognition still works, just slower.

Linux

1. Download Mockingbird.AppImage.
2. Make the file executable and run it:

chmod +x Mockingbird.AppImage
./Mockingbird.AppImage

First launch

1. Check the device indicator in the toolbar: GPU or CPU.
2. Upload a PDF résumé on the Résumé tab (optional).
3. Press Start and speak — the draft appears on the Interview tab.

Updating

The app checks for new releases on startup. Manually — on thereleases page.

Next: choosing an STT model andoperating modes.