A LITTLE CAT FOR YOUR READING

Select text.
Stay curious.

Keep the sentence in view. Explain what it means in plain Traditional Chinese, or translate it directly, in a small window near your selection. Text stuck in a video or image? Press ⌥S and frame it.

Download free View source ↗

v0.2.0 · macOS 14+ · Apple Silicon
Developer ID signed and Apple notarized. Your own Gemini API key and internet connection are required. App UI is in Traditional Chinese.

The orange and cream UTUVO cat iconActual UTUVO Explain window explaining a WebAssembly sentence in Traditional Chinese
Actual app · Sample sentence
THREE WAYS INExplain / Translate / Frame
NATIVE MAC APPSwiftUI + AppKit
YOUR CHOICEMIT open source · Your API key
01 / HOW IT WORKS

Keep reading.
Stay in place.

One shortcut from selection to answer. If another app does not expose its selected text, you can paste into Explain directly.

01

Select a sentence

Highlight text in your browser or another Mac app.

02

Press a shortcut

⌥D explains; ⌥⇧D translates. You can also switch modes in the window.

03

Read nearby

The compact window tries to appear near your selection, so you can continue reading.

02 / TWO WAYS TO READ

Understand it.
Or translate it.

The two modes produce different results and the window always labels the current one.

⌥D

Plain-language explanation

Explains the meaning in Taiwan Traditional Chinese, adding essential context when useful.

Actual app screenshot in explanation mode
⌥⇧D

Translation

Other languages to Taiwan Traditional Chinese; Chinese to English. The result contains only the translation.

Actual app screenshot in translation mode
03 / FRAME THE SCREEN (NEW IN 0.2)

Text in a video
or image? Frame it.

Subtitles, screenshots and scanned PDFs don't let you select text. Press ⌥S, drag around the words, and Explain reads them on your Mac, then explains or translates.

Tutorial video (1:37, Traditional Chinese narration by AI voice): setup, selecting text, and framing the screen.
  1. 01

    Press ⌥S and frame the text

    A crosshair appears. Drag around the words. Press Space to pick a whole window, or Esc to cancel. Press ⌥⇧S to go straight to a translation.

    Illustration of framing an English subtitle in a video window
  2. 02

    Release — the window appears

    It shows “recognizing” first, then the explanation. The recognized text sits on top; fix any typo with Edit and send it again.

    UTUVO Explain window next to a video, explaining the subtitle
  3. 03

    Switch to a translation

    Flip between Explain and Translate for the same text, and copy the result with one click.

    UTUVO Explain window showing a Traditional Chinese translation of the subtitle
  4. 04

    First time: allow Screen Recording

    Framing needs macOS “Screen & System Audio Recording”. macOS asks the first time you press ⌥S; turn it on and relaunch. A built-in tutorial is always one click away in the menu bar.

    Page three of the built-in UTUVO Explain tutorial

Recognition runs on your Mac and the captured image is deleted right after; only the recognized text is sent to Gemini. The video frame shown is a demo.

04 / YOUR SETUP

Your key.
Your pace.

Explain does not require a new account. Add your own Gemini API key in Settings. The selected text is sent to Google only when you trigger an explanation or translation. Google controls its free-tier limits and data-use terms.

  • Key in macOS KeychainIt is not embedded in the source code or saved as a translation history.
  • One macOS permissionAccessibility access reads the current selection and enables global shortcuts. You can paste text manually when needed.
  • No history listThe app does not retain explanation or translation history. Check whether a passage is suitable to send to Gemini before using it.
GOOD TO KNOW

Before you begin.

Is the app free?

The app is free and MIT licensed. You need your own Gemini API key. Google's free-tier limits, rate limits and data-use rules can change; see the official pricing page.

Why does it need Accessibility access?

The app uses macOS Accessibility to read the current text selection and listen for its global shortcuts. It does not log keystrokes. You can paste text manually if another app does not expose its selection.

Does framing upload my screen?

No. The framed area is read on your Mac with Apple’s text recognition and deleted right away. Only the recognized text is sent to Gemini, when you ask. Framing needs Screen & System Audio Recording permission and runs only when you press ⌥S.

Which Macs are supported?

The current installer targets macOS 14 or newer on Apple Silicon. Source is available to build yourself; Intel Macs have not been verified.

SMALL TOOL. CLEAR WORDS.

Understand the next sentence.

Start with a selection. Keep your place.

Download UTUVO Explain ↓