Step 1 — find the caption band
Scrub through the clip and note where the text sits. Most hardsubs occupy a horizontal strip in the lower third to lower half of the frame, and stay in the same place for the whole video.
If the position changes mid-clip, or a second element appears later (a watermark, a lower-third), plan for two passes: one for the caption band, one for the other element.
Step 2 — pick automatic or manual
Automatic detection is tuned for white, centred captions in the bottom half of the frame. It is the fastest option and usually the right one for dialogue subtitles.
Manual selection is for everything else: outlined captions, coloured text, captions positioned high in the frame, or a logo you also want gone. Draw a box around the text and the model restricts the repair to that area.
Step 3 — check the result before exporting everything
Run a short clip first. Look at the repaired area while the video plays, not on a still frame — flicker and shimmer only show up in motion, especially over textured backgrounds like fabric, hair or foliage.
If the area looks soft, switch to the higher-quality erase mode. It costs more credits per minute but reconstructs texture instead of smoothing it.
Frequently asked questions
What is the difference between hardsub and softsub?+
Softsubs are a separate text track the player renders on top of the video. Hardsubs are rendered into the frames at export time. Softsubs can be toggled or replaced instantly; hardsubs must be removed pixel by pixel.
Can I remove hardcoded subtitles for free?+
Every new account gets one free run, enough for a 15-second clip. Longer videos use credits — one credit is one minute of standard-quality erase.
Will the audio be preserved?+
Yes. Only the picture is modified; the audio stream is copied through untouched.
Does removing subtitles reduce quality?+
The output is re-encoded, so there is a small generational loss. Resolution and frame rate are preserved, and the erased area is reconstructed rather than blurred.