TikTok Transcript API
Turn a public TikTok link or uploaded MP4 into structured JSON: spoken transcript, burned-in hook text, summary and chapters over one REST endpoint.
https://scribiz.com/api/v1/contextQuickstart
Speech and on-screen text in one call
Send a public TikTok URL (or an uploaded file id) to /api/v1/context with mode: "both", then poll /api/v1/jobs/:id or stream SSE events.
JOB=$(curl -s https://scribiz.com/api/v1/context \ -H "Authorization: Bearer $SCRIBIZ_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{"url": "https://www.tiktok.com/@creator/video/7345678901234567890", "mode": "both"}' | jq -r .job_id)[ -z "$JOB" ] || [ "$JOB" = "null" ] && exit 1while true; do STATE=$(curl -s "https://scribiz.com/api/v1/jobs/$JOB" \ -H "Authorization: Bearer $SCRIBIZ_API_KEY") STATUS=$(printf '%s' "$STATE" | jq -r .status) case "$STATUS" in succeeded) break ;; failed|canceled) printf '%s' "$STATE" | jq .error; exit 1 ;; esac sleep 3doneprintf '%s' "$STATE" | jq '{segments: .result.transcript.segments, scenes: .result.visual.scenes}'{ "job_id": "job_2DUafpGltJTqVKlAYlIREm", "result": { "meta": { "durationSeconds": 24, "title": "Fold a fitted sheet in 3 steps" }, "source": { "kind": "url", "url": "https://www.tiktok.com/@creator/video/7345678901234567890" }, "synthesis": { "summary": { "tldr": "Shows a three-step method for folding a fitted sheet by tucking opposite corners together." } }, "transcript": { "language": "en", "segments": [ { "end": 3.4, "start": 0, "text": "Nobody can fold a fitted sheet. Watch this." }, { "end": 7.8, "start": 3.6, "text": "Tuck each corner into the next one, like gloves." } ] }, "visual": { "scenes": [ { "description": "Creator holds up a crumpled fitted sheet.", "end": 4, "onScreenText": [ "Fold a fitted sheet in 3 steps" ], "start": 0, "timingApprox": true } ] } }, "status": "succeeded"}Audio + visual frame OCR vs caption-only scrapers
What a caption scraper misses on TikTok: burned-in hooks and silent clips.
Evaluating developer APIs? See our side-by-side comparison on Scribiz vs Supadata.
Choose speech, screen OCR, or both per request
mode: "both"
Transcribes spoken audio and reads burned-in hooks, overlays, slides and end cards from the frames onto one timeline.
mode: "listen"
Spoken transcript with timestamps, speakers when there is more than one, summary and chapters when you do not need frame OCR.
mode: "watch"
Reads only on-screen text and visual scene notes. For silent tutorials, recipes and music-only hooks.
Every free account starts with 30 minutes a month across the REST API, CLI, web app and MCP server. Opening a stored result again is free. See pricing and sources and limits.
Questions & answers
Common questions about the Scribiz TikTok Transcript API.
- Send a POST request to https://scribiz.com/api/v1/context with your Bearer API key, the public TikTok URL, and mode: "both" (speech + burned-in screen text) or "listen" (speech only). Poll GET /api/v1/jobs/:id until status is "succeeded".
- Yes. With mode: "both" or "watch", Scribiz samples frames and returns visual.scenes with approximate scene times, description and onScreenText alongside transcript.segments.
- Public TikTok links are fetched on a best-effort basis. When TikTok rate-limits or blocks a URL, upload the MP4 or extracted audio (up to 100 MB) via PUT /api/v1/uploads and pass upload_id to POST /api/v1/context.
- Listen uses 1 minute per minute of video and Both uses 2 minutes per minute. A 30-second TikTok uses about 1 minute in Both mode. Free accounts include 30 minutes a month, and opening a stored result again is free.
- Yes. Connect https://scribiz.com/mcp with the same API key in Claude, Cursor, VS Code or Codex to call get_video_context, search_video and get_transcript directly.
Start with 30 free minutes.
Get a free API key in your dashboard or connect the remote MCP server inside Claude, Cursor or Codex.


