Support

Getting started, system requirements, and frequently asked questions.

Getting started

1. Download:Get the latest release from GitHub releases.

2. Run:Windows: run the installer. Mac: unzip, Control-click RotaAI.app, choose Open. Linux: chmod +x RotaAI.AppImage && ./RotaAI.AppImage

3. Onboarding:Pick your transcription backend (Groq, Gemini, or Ollama). Add your API key if using cloud.

4. Dictate:Press F9 in any app. Speak. Release F9. Your text appears.

System requirements

Windows 10/11, macOS 13+, or Linux (Ubuntu 20.04+, Fedora 36+, Arch)

RAM

4 GB minimum, 8 GB recommended

GPU

Optional. NVIDIA with 4GB+ VRAM for local transcription

Disk

~600 MB for dependencies + model size (140MB–3.1GB)

Microphone

Any built-in or external mic

Frequently asked questions

Is this really free?

Yes. MIT licensed open source. Groq and Gemini both have free tiers. Ollama is completely free. No pro plan, no premium tier, no credit card needed.

How does the download work?

Windows: run the installer. Mac: unzip RotaAI-macOS.zip, Control-click RotaAI.app, choose Open, and grant permissions on first launch. Linux: download the AppImage, mark it executable (chmod +x RotaAI.AppImage), and run it.

Can I use it without internet?

Yes. Install Ollama, download a Whisper model (small is 480MB), and Rota works 100% offline. No API keys, no accounts, no internet.

What about Mac or Linux?

Windows, macOS, and Linux builds are published from GitHub releases. The macOS build is currently unsigned, so first launch uses the standard Control-click Open workaround until Developer ID notarization is available.

System requirements?

Windows 10/11, macOS 13+, or Linux (Ubuntu 20.04+, Fedora 36+, Arch). 4GB RAM minimum, 8GB recommended. For local GPU: NVIDIA GPU with 4GB+ VRAM. CPU-only works on any modern quad core.

Which Whisper model should I use?

Base (140MB) for speed. Small (480MB) for best balance. Large v3 turbo (1.5GB) for best accuracy if you have 4GB+ VRAM.

How do I get a Groq API key?

Go to console.groq.com, sign up (free), and create an API key. Paste it into Rota during onboarding.

Can I use my own Whisper model?

Yes. Rota supports loading custom model paths. Point it to your fine-tuned model in settings.

Still need help?

Open an issue on GitHub. I read every issue and try to respond quickly.