Translate the on-screen text in your videos, not just the audio

The challenge

Today when you translate a video, only the spoken audio track is translated — the text shown on screen isn't touched. For many videos that's fine, but for software training and how-to content the on-screen texts and labels are a core part of what's being taught. Teams often have their source material in one language (for example German) while colleagues work in English and Spanish, so a fully translated version really matters. The only way around it today is to record the software interface in English so that just the voiceover needs translating — a workaround, not a real answer.

Our approach

Let's have the AI editing agent also recognize the text elements shown on screen and translate them, so a translated video covers both the voiceover and the on-screen texts and labels — not only the audio.

Expected impact

• Fully translated training and how-to videos, including the on-screen texts and labels

• Reach colleagues and audiences across languages without redoing the recording

• Drop the workaround of keeping the interface in one language just to make translation work