FAQ

What is Amical?

Amical is a Voice AI app for everyday tasks — dictation, note-taking, meeting transcription, and more. It's context-aware, so it formats your speech differently depending on whether you're writing an email, chatting on Slack, or taking notes.

How does it work?

Amical combines speech-to-text AI models (like Whisper) with context-aware formatting. It detects your active app, transcribes your speech, applies formatting rules appropriate to the context, and pastes the result directly where your cursor is.

Is Amical free?

Yes. Amical is free and open-source under the MIT license. The source code is available on GitHub.

Is my data private?

Amical is open-source and supports local AI models for fully offline transcription. You can also bring your own API keys for cloud models. Your data stays under your control.

How accurate is the transcription?

Amical uses state-of-the-art Whisper models for transcription and LLMs for post-processing — removing filler words, fixing grammar, and formatting output. Accuracy depends on the model you choose. Larger local models and cloud models deliver the highest accuracy.

What platforms are supported?

• Mac and Windows — Available now. Download from the downloads page or install via Homebrew on Mac: brew install --cask amical
• iOS and Android — In private beta. Join the beta.
• Linux — Planned.

How do I install Amical?

On Mac, the fastest way is Homebrew:

brew install --cask amical

Or download directly from the downloads page. See the Quick Start guide for full setup instructions.

What languages are supported?

Amical supports 100+ languages through Whisper. See Multi-Language Support.