Skip to main content

​
Why subtitles aren’t translated in Visual Translate

Subtitles are generated from the spoken audio or voiceover in the video, so they are inherently audio-based and tied to speech content. Visual Translate, on the other hand, is designed to translate explanatory visual text that appears directly in the video frame. This includes elements such as labels, titles, and slide content—text that visually explains or supplements the video content, rather than representing spoken language. For this reason, Visual Translatedoes not process subtitles by design.

​
Why we separate subtitles from visual text

​
Subtitles require audio context

Subtitle translation depends heavily on spoken context, sentence flow, and timing.
Visual Translate works on isolated visual text and does not have access to the full linguistic context needed for high-quality subtitle translation. If subtitles were translated through Visual Translate, the results could be inaccurate or unnatural.

​
Avoid duplicated or conflicting results

Subtitles are better handled through audio-based workflows:
  • Translate Subtitles
  • Translate & Dub
If subtitles were also translated visually:
  • You might get duplicated subtitles. And text-based subtitle translations could conflict with audio-based translations.
  • When dubbing is applied later, the subtitle text and the spoken dubbing may differ, as they are generated from two separate translations—one visual and one audio-based.

​
Which tool should you use next?

After completing Visual Translate, if you still want to translate the subtitles that haven’t been processed yet, here’s what to do next:
  • If you want to translate subtitles only (no dubbing)
    → Export the video and use Translate Subtitles.
  • If you want both dubbing and subtitles in the new language
    → Export the video and use Translate and Dub, then turn on subtitles.
Currently, Translate Subtitles and Translate and Dub do not support erasing existing subtitles. These tools can only add translated subtitles on top of the video.If you need to temporarily cover the original subtitles, you can add a subtitle background to mask them.We’re actively working on enabling subtitle erasing across all related tools.
Ai Translation Other Tools

​
What’s coming next

We’re working toward an all-in-one translation workflow. In the future, you’ll be able to upload your video once, select all the elements you want to process—such as visual text, subtitles, dubbing, and lip sync—and complete the entire translation in a single workflow. Stay tuned.