Captions from TikTok, Reels, Shorts and YouTube are usually burned into the video. Draw a box, erase them with AI, and export a clean clip — free to start, no watermark on the result.
Manual translation, hired voice actors, days of waiting. Creators and teams lose time and money on every single video.
Skip the old workflowThe auto-captions on a reposted TikTok sit right where I want my own subtitles to go.
Client footage came back with the previous editor's handle burned in as a caption overlay — I need it gone.
I want the words for a transcript, but I also want a clean copy of the video without them.
Upload, pick languages, and let the AI handle the rest. No software to install.
A few clicks, that is all.
A few clicks, that is all.
A few clicks, that is all.
A few clicks, that is all.
Caption is just another word for the same problem: text that the creator rendered onto the video instead of attaching as a subtitle file. On social platforms it usually arrives that way by default — auto-captions on TikTok, Reels and Shorts, the big animated subtitles that fill half the screen, or a lower-third label that stays put for the whole clip. None of it can be turned off afterwards, because it is already part of the picture. SpeakVid erases it with a box: mark the caption area, choose the strength, look at the before/after preview, export. If you actually want the words — for a repost with better styling, or for a transcript of a client's video — run the transcription tool first and keep the SRT, then come back and erase the burned-in version. The output has no watermark added, and the area outside your selection is not re-encoded, so the rest of the frame keeps its original quality.
Full-width caption bars, one-word-at-a-time pop text, and handle watermarks in the same frame are all normal inputs here — box them together in a single task.
If the caption text is useful, transcribe the clip first and download the SRT — then erase the burned-in layer. You get both the text and a clean video from the same upload.
Paste a TikTok, Reels, Shorts or YouTube link instead of downloading first. Works in the browser with nothing to install, and the export carries no added watermark.
Sign up for 20 credits, upload the video or paste a link, box the caption area, set the strength and export the cleaned MP4. Erasure runs at 1 credit per 30 seconds of footage, so the free allowance is roughly 10 minutes of video. Failed tasks are not charged.
Yes, those are exactly the case it is built for. Auto-captions are burned into the frames, so they cannot be switched off in the player — they have to be treated on the pixels. Box the caption band once and the same area is cleaned for the entire clip.
1 credit per 30 seconds of video, and the export carries no watermark added by SpeakVid. You can also run the before/after preview to check the result on real frames before spending anything on the full task.
Yes. Run the transcription tool on the same clip first, download the SRT, then erase the burned-in captions here. That gives you a transcript for repurposing and a clean video for re-editing, from one upload.
It depends on what was behind the text. A caption over a static background or a smooth gradient cleans up almost invisibly; one sitting on a busy, fast-moving shot will look smoother inside the selection than around it. The before/after preview shows you exactly which case you are in before you export.
We use local storage only to keep you signed in — no tracking cookies, no third-party ads. See Privacy Policy
SpeakVid does not use tracking or advertising cookies. The only storage used is listed below.