Get the spoken script out of any TikTok so you can study the hook, see the exact claims, build a swipe file and write your own version. Paste the link, get timestamped text.
Open JotVox →The words that were said, in order, with timestamps. Not a caption file to burn back onto a video, the script itself, as text you can read on one screen and paste into a doc.
That distinction matters because of what happens next. A caption file goes back into an editor. A script goes into analysis: where the hook lands, how long the setup runs, what the claim is, how the close is worded.
If all you want is a video's words for subtitles or notes, the TikTok transcript generator is the page for that job. This page is about taking a script apart and reusing it.
Timestamps make the first three seconds measurable instead of a feeling. Pull the script, look at what sits before the 00:03 mark, and the structure is plain:
Do that across ten videos in one niche and the pattern stops being a guess.
Yes, and it is the most practical use of the whole thing. Spoken claims in a sixty-second product video go past too fast to note by ear. In text they sit still.
People use that for competitive research, for checking what a creator promised about a product, for a compliance read before a brand runs something similar, and for lifting the specific wording an audience responded to. Every line has a timestamp, so a claim can be pointed at rather than described.
One thing to hold in mind: a claim said out loud and a claim flashed on screen are not the same thing here. Only spoken audio and real caption tracks end up in the text, which the last section covers.
Work in batches. Save a video, extract its script, export to TXT, and keep the files in one folder with the view count in the filename. After thirty or forty, you have something you can read across. Nothing to install either, since it all runs in the browser.
Exports are TXT, SRT, VTT and MP3. TXT is the one you want for a swipe file. Add the AI summary if you would rather have key takeaways and quotes pulled out for you, which uses credits as transcription does.
Adapting a TikTok script for Reels or Shorts starts the same way: get the text, keep the hook structure, rewrite the pacing and references for the other platform. Related: repurposing video transcripts and how to transcribe a TikTok video.
One limit to be clear about before you rely on this: on-screen text is not extracted. Burned-in captions, text overlays and title cards are pixels in the video frame, not data. JotVox reads spoken audio and any real caption track. If a video makes its point purely through text on screen, there is nothing to pull.
Other rough spots:
When a TikTok carries no caption track, AI transcription handles the audio at 98%+ accuracy across 99 languages. See getting text from a video with no captions.
1. Copy the video link Open the TikTok, tap Share and copy the link. Any public video works and you do not need a browser extension. If you have the clip saved as a file instead, upload it from your device, Google Drive or Dropbox.
2. Paste it and get the script Drop the link into JotVox. If the video carries a real caption track, that text is grabbed free with no sign-up. If it has none, AI transcription runs on the audio using one credit.
3. Mark the hook Read everything before the 00:03 timestamp and write that opening line out separately. It is the part worth studying, and the part you will rewrite for your own video.
4. Export and file it Export to TXT and save it into a swipe folder, named so you can tell which videos performed. Run the AI summary if you want takeaways and quotes separated out for you.
Paste the video link into JotVox and it returns the spoken script with timestamps. If the TikTok has a real caption track, that text is pulled free with no sign-up. If it has none, AI transcription runs on the audio using one credit, at 98%+ accuracy across 99 languages. You can also upload a saved clip from your device, Google Drive or Dropbox.
No. Text overlays, burned-in captions and title cards are pixels in the video frame, so there is nothing to extract from them. JotVox works from spoken audio and from any genuine caption track attached to the video. If a TikTok delivers its message entirely through on-screen text with no narration, you will get little or nothing back.
The output is similar text, but the job is different. A transcript generator is for getting a video's words as captions or notes, and the TikTok transcript generator page covers that. This page is for creators studying scripts: isolating the hook, reading the claims line by line, building a swipe file and adapting a script that worked for another platform.
Yes, and the transcript is the right starting point. Extract the script, keep the hook structure and the order of the beats, then rewrite the wording for the other platform's pace and audience. Copying a script line for line rarely transfers well, because delivery speed and references differ per platform. What transfers is the structure, not the exact sentences.
A clip with no speech produces no script, because there is nothing said to transcribe. Music sitting loudly over a voice lowers accuracy, as does very fast or sped-up delivery, so expect to fix a few words in those cases. Videos with clear narration come back clean. Where a caption track exists, JotVox uses that instead, which sidesteps audio problems entirely.