Video sound, screen text or a live speaker: which iPhone translation mode should you use?

Start with what you want to understand

A Japanese livestream with no subtitles is a listening problem, even if it is playing on a screen. A Japanese game menu is a reading problem, even if the game also has sound. A lecturer at the front of a room is a microphone source. Picking the right mode matters more than the name of the app you are watching.

Sound: keep watching in the app you already use

Sound is the starting point for YouTube, TikTok LIVE, Weverse, Twitch and videos in Safari. Choose the spoken language and the language you want to read, then start sound translation and confirm the iOS sharing sheet. Tap the empty space below the sheet, wait for the subtitle window, and return to the video. It receives playback audio rather than listening to your speaker through the microphone.

The floating card can show the original, the translation or both. If the video app opens its own Picture in Picture, it may take over the one window iOS allows. Protected content or a source that does not share audio may not work; use the free trial with your usual content first.

Screen: read the words visible in another app

Screen reads visible text through screen sharing and shows its translation in the floating window. Use it for a web page, a PDF paragraph or a game menu. Stop scrolling and let the text settle. It does not rewrite the original app, replace the text inside a comic bubble or create a translated PDF file.

Live: follow a speaker nearby

Live uses the iPhone microphone. Set the spoken and target languages, allow microphone access and place the phone near the speaker. It is useful for a lecture or a conversation when you want written translation. It is not a meeting bot, does not join a call, and cannot make a distant or noisy voice clear.

One app, three inputs: what to prepare

Use an iPhone running iOS 26 or later. Download the required language models before you need them; processing can then run on the device. Available languages depend on the mode, device and installed models. Sound and Screen cannot run together. Saving translation records is a Premium feature, separate from recording a video clip.

Questions

Which mode should I use for a game with spoken dialogue?

Use Sound for spoken dialogue and Screen for written menus or dialogue boxes. Stop the current mode before switching; they do not run simultaneously.

Does Screen translate speech with no written captions?

No. Screen translates visible text. If there is only speech, choose Sound for playback or Live for the microphone.

Do I need three different apps?

Hono Translate includes all three modes. Choose the one that matches the source; each has its own permissions and limits.

Choose your next step

Guides for specific apps

View other guides