TikTok Transcript Generator
Turn a TikTok video into text, SRT or VTT subtitles — with timestamps. Everything runs on this device; your file is never uploaded.
Drop a video or audio file here
or
MP4 · MOV · WEBM · MP3 · WAV · M4A · OGG · AAC · FLAC — any video or audio, not just TikTok
Download the video, then . It's the same two taps, and nothing gets uploaded.
Preparing…
What this tool does
This is a TikTok transcript generator, and also a general TikTok video to text converter. It reads the spoken audio in a video or audio file and writes it out as plain text with timestamps. From the same transcript you can export SRT or VTT subtitle files, ready to drop into CapCut, Premiere Pro, or DaVinci Resolve.
It is not limited to TikTok. Any video or audio file your browser can decode works — TikTok video transcription, a screen recording, a podcast episode, a voice memo. The engine is OpenAI's Whisper, running locally through WebAssembly.
How to generate a TikTok transcript
- Get the file. In the TikTok app, tap Share, then Save video — or use whatever way you normally keep the clip. Any video or audio file on this device works.
- Drop it in. Drag it onto the drop area above, or pick the model you want and hit Transcribe.
- Read or export. Copy the text, or switch to SRT or VTT and download it.
Why the link box doesn't fetch the video
Several tools accept a TikTok URL and return a transcript. They can do that because they run a server: the server downloads the video for you, then transcribes it. This page deliberately has no server — that is the whole point. A URL box that looks like it fetches, but quietly sends your link to someone's backend, would be worse than an honest explanation. So the link box tells you the truth and points you at the file drop.
The trade-off is real and worth stating plainly: you have one extra step (saving the video), and in exchange your file never leaves this device. No upload, no queue, no retention policy to read.
Supported input formats
Anything your browser can decode. In practice that means the common containers: MP4, MOV, WEBM, MP3, WAV, M4A, OGG, AAC, FLAC. Video files are the normal case — audio is extracted in the browser before transcription.
Supported languages
Multilingual models (whisper-tiny, whisper-base, whisper-small) auto-detect the spoken language and transcribe in it. Whisper's published language list covers 99 languages. The whisper-tiny.en model is English-only.
Source for the 99-language figure: the Whisper model card published by OpenAI (openai/whisper on Hugging Face), which lists the supported languages and their language codes. Common ones include English, Spanish, French, German, Portuguese, Italian, Japanese, Korean, Chinese, Arabic, Hindi, Russian, Turkish, Vietnamese, Thai and Indonesian.
Accuracy
This page does not print an accuracy percentage. Whisper's word error rate depends heavily on audio quality, accent, background music and speaking speed, so a single headline number would be marketing rather than information. What can be said precisely: the models offered here are the official Whisper checkpoints converted to ONNX by the Xenova project — tiny, base and small — and larger models are generally more accurate. If a transcript matters, use small.
FAQ
Is anything uploaded to a server?
No. The file is read and decoded in the browser, and the Whisper model runs in the browser via WebAssembly. There is no backend to upload to. The only network request is the one-time download of the model weights.
Why is the first run slow?
The model weights are fetched once and cached by the browser. whisper-tiny is the smallest; whisper-small is much larger and noticeably slower, especially on a low-memory machine. After the first run, the cached model loads quickly.
Can I paste a TikTok link instead?
You can paste one, and the box will tell you what to do next — but this page will not fetch the video, because it has no server to do that with. Save the video and drop the file in.
Does it work on a phone?
The layout adapts down to phone widths. Be aware that running Whisper in a mobile browser is heavy — on a phone, prefer whisper-tiny and short clips.
What are TXT, SRT and VTT?
TXT is plain text with timestamps at the start of each line. SRT and VTT are subtitle formats: numbered cues with start and end times, accepted by most video editors and by YouTube and Vimeo upload.
Is there a file size limit?
No limit is imposed by this page. The practical ceiling is your device's memory — transcription loads the whole audio into memory, so very long recordings on a low-memory machine may fail. Splitting the file is the usual fix.
Privacy
Nothing you drop here is transmitted anywhere. There is no analytics, no cookie, no account. Close the tab and it's gone.
Working with Instagram Reels too
If the clip came from Instagram rather than TikTok, the same engine is set up for it on a dedicated page: the Instagram transcript generator handles Instagram Reel transcript and Instagram video to text the same way — in-browser, no upload.