Introduction

Nexo Whisper turns your voice into text in any macOS application, instantly. Press a shortcut, speak, release, and the text appears wherever your cursor was.

It's native, fast, and respects your privacy: transcription runs locally on your Mac with Whisper models. If you want to improve the text with AI (correct, format, summarize), you connect your own API key or use your already-installed CLI — we don't intermediate.

What you can do

• Dictation in any app — Slack, Notion, Mail, Cursor, browsers.
• Local processing — Whisper models run on your Mac, no audio sent to the cloud.
• 99 languages supported by Whisper.
• 8 enhancement prompts bilingual (Email, Coding, Formal, Summary, Fun, etc).
• App Modes — different configurations per active app.
• Personal dictionary for technical terms and proper nouns.
• Global shortcuts configurable (push-to-talk or toggle).

How it works

1. Press Fn (configurable in Settings → Shortcuts).
2. Speak.
3. Release Fn.
4. The text is automatically pasted where your cursor was.

If AI enhancement is active, the text passes through the LLM you configured before being pasted — adding 1–3 seconds depending on the provider and model.

Privacy

Nexo Whisper does not send your audio to any server. Everything runs on your Mac. The only exception is if you activate AI enhancement using a cloud provider (Anthropic, OpenAI, Gemini, Groq) — in that case the transcribed text (not the audio) is sent to the API of the provider you chose with your own key.

Next steps

• Installation — Download and get started.
• Permissions — Grant the 4 system permissions.
• Recommended models — Which one to choose for your Mac.