Listen

1. Open Listen. Choose the audio language — English, Chinese, Japanese, Korean or Russian — and the subtitle language.
2. The first time you use a language, click Prepare speech component next to Start captions. It downloads that language’s recognition model, about 170–550 MB depending on the language (Chinese, Japanese and Korean share one).
3. Under Caption content, choose Translate, Transcribe (the original words only — the fastest) or Subtitle book.
4. Click Start captions.

Recognition runs on your PC; your audio is never uploaded. Each sentence is translated as a whole, once it’s finished.

Caption appearance

The number of lines, size, shadow and backing are in Caption appearance. While it’s open, two sample captions show on screen, so you see every change right away. To hear the translations, turn on Read captions aloud. The audio source and a captions on/off key are in More listening settings.

Subtitle book

Have the video’s subtitle file (.srt, .vtt, .ass and others)? First pre-translate it into a glossary. Then pick that glossary and choose Subtitle book: NativeLens only listens to keep time and shows the ready-made translations. If it can’t stay in sync, it asks you to check the audio language and the subtitle file, and then falls back to Transcribe.