What you get back
Short paragraphs, each marked with the second it starts at. Click to jump to a line, copy
a quote with its timecode, or export the lot. .txt keeps the timings inline;
.srt and .vtt give you the same words as a subtitle file you can
drop into an editor.
We tested it on a 31-second clip with a single narrator. It came back as seven timestamped segments, language detected as English, and the text read as clean, complete sentences. From pasting the link to finished text it took 72 seconds on the first run and 94 on the second, which covers any queueing as well as the transcription itself.
The engine is OpenAI Whisper, large-v3-turbo. If you plan to lean on the output in a language other than English, read how accurate Whisper actually is first. The gap between languages is wider than you might expect.
Which links work, and which can't
- One video: yes. The long address with
/video/in it, or the short share link from the Share button. Both forms are accepted, and the tracking parameters can stay on. - A profile: no.
tiktok.com/@namepoints at a list. Open the video you want and paste that link instead. - A sound page: no. It lists every video that uses one track, so there is no single recording to transcribe.
- Private or deleted videos: no. If the link needs a login to watch, it needs one to fetch, and we will not ask for yours.
Why bother turning a 30-second video into text
A TikTok is easy to watch and awkward to reuse. You cannot search it, quote it accurately, or hand it to someone who would rather read. A transcript solves all three. People mostly use it to pull an exact quote for an article without typing it out from a paused screen, to keep a written copy of advice they saw once and will not find again, to check what a creator really claimed before repeating it, and to get the script of their own videos back out for reuse.
It also helps accessibility. A video that carries its meaning only in speech is closed to anyone who cannot hear it. The .srt and .vtt export turns the same transcript into captions you can load into a video editor.
Short clips, different limits
The free tier stops at 75 minutes of audio per transcript, which matters for a podcast and never for a TikTok. The limit you will actually meet is the other one: fifteen transcripts a day from one address. Three links at once run as a batch, so a handful of saved clips is one paste, not several.
Speed depends on who else is using the service. A clip this short is quick on its own, but a busy hour adds waiting, and the page shows your place in the queue instead of leaving you guessing. If you need it faster, there is a paid fast lane, $5 for 10 hours of audio, which skips the queue.
When a link won't go through
There are three usual causes. The video is private. It was deleted by its author. Or the link is not a video at all, but a profile or a sound page. If a public video you can open in a browser still fails, try it once more before assuming the link is the problem.
Looking for another source? There are pages for Instagram reels, YouTube videos, podcast episodes and direct audio links. If the clip is not in English, it can come back translated, and on a phone the whole thing runs from the share sheet.
Frequently asked
Which TikTok links work?
A link to one video. That is the long address with /video/ and a number in it, or the short link TikTok gives you from the Share button (vm.tiktok.com, vt.tiktok.com or tiktok.com/t/). Paste it whole; the tracking parameters on the end do no harm.
Can I paste a profile link and get every video?
No. A profile is a list, not a recording, and each video needs its own link. If you have several saved, paste up to three at once and they run as a batch.
Do I need a TikTok account?
No account, and nothing to install. A video that anyone can open in a browser can be transcribed. A private video, or one that has been removed, cannot, and you get an error instead of a transcript.
What if the video is only music?
Then there is no speech to write down, and the transcript will be short or empty. The tool transcribes what is said. It does not identify songs, and a sound page (the one that lists every video using a track) is not a video, so it is not accepted.
What language does it expect?
None. It detects the language itself, out of 100 supported, so a clip in Spanish or Tagalog is handled the same way as one in English. Accuracy still varies by language: it follows how much of each one the model has seen.
Can I get subtitles rather than plain text?
Yes. Every transcript can be saved as .txt, .srt or .vtt, and the two subtitle formats carry the same timings as the text.
What does it cost?
Nothing, and there is no signup. The free tier allows fifteen transcripts a day from one address, each up to 75 minutes of audio, which a TikTok will never get near. If you need to run a lot of clips quickly, there is a paid fast lane: $5 for 10 hours of audio.
The link is already on your clipboard.
Paste it above. Need subtitles instead of notes? The caption export is the same transcript in a different wrapper.