TikTok Transcript Scraper — Captions + Whisper Speech-to-Text

TikTok Transcript Scraper turns TikTok videos into text, without a TikTok account or login. For each video it returns the full transcript, timed segments and optional SRT subtitles, plus the video's views, likes, comments, shares, saves, hashtags, author and publish date. It first uses TikTok's own captions (fast and cheap) and, when a video has none, transcribes the audio with built-in Whisper speech-to-text in 90+ languages. Paste video URLs, short links, or creator usernames to transcribe their latest videos.

What you get

• TikTok captions first: instant, accurate, and the cheapest option.
• Whisper speech-to-text fallback for videos without captions (tiny, base or small model), auto-detecting 90+ languages.
• Timed segments and ready-to-use SRT subtitles for every transcript.
• Video stats included: views, likes, comments, shares, saves, hashtags, music, author and publish date.
• Inputs: video URLs, vm.tiktok.com short links, or usernames (latest 10 videos per creator).
• Fair billing: videos without speech and failed videos are free; speech-to-text is charged only when it runs.

How to use TikTok Transcript Scraper

1. Paste TikTok video URLs (or short links), and/or Creators (usernames like @nbcnews) to transcribe their latest videos.
2. Keep Transcript source on TikTok captions, then speech-to-text, or choose captions only (cheapest) or speech-to-text only.
3. Optionally set a Language (e.g. en) and turn on SRT subtitles.
4. Run the Actor and export transcripts to JSON, CSV or Excel, or send them to Google Sheets, Make, Zapier or your AI pipeline.

Input example

{
  "profiles": [
    "@nbcnews"
  ],
  "maxVideosPerProfile": 2,
  "includeSrt": true
}

All fields are described in the Input tab.

Output example

{
  "videoId": "7699551142543445262",
  "url": "https://www.tiktok.com/@nbcnews/video/7699551142543445262",
  "authorUsername": "nbcnews",
  "authorName": "NBC News",
  "description": "David Solomon, CEO of Goldman Sachs, speaks with NBC News' Christine Romans on his routine…",
  "hashtags": [],
  "createdAt": "2026-10-22T20:15:00+00:00",
  "durationSeconds": 104,
  "views": 8343,
  "likes": 210,
  "comments": 15,
  "shares": 9,
  "saves": 20,
  "musicTitle": "original sound - nbcnews",
  "coverUrl": "https://p16-common-sign.tiktokcdn-us.com/example.jpeg",
  "transcript": "Hmm. That's good. Grandma's always have the best advice. She was a wise woman. The first one is, what are the three things you do before 7:00am? Wake up…",
  "wordCount": 367,
  "language": "eng-US",
  "transcriptSource": "tiktok_captions",
  "whisperModel": null,
  "captionLanguages": [
    "eng-US"
  ],
  "segments": [
    {
      "start": 0.1,
      "end": 1.26,
      "text": "Hmm. That's good."
    }
  ],
  "srt": "1\n00:00:00,100 --> 00:00:01,260\nHmm. That's good.\n",
  "note": null,
  "scrapedAt": "2026-10-23T04:30:00+00:00"
}

Output fields

Field	Description
videoId	TikTok video ID.
url	Video URL.
authorUsername	Creator's username.
authorName	Creator's display name.
description	Video caption text.
hashtags	Hashtags on the video.
createdAt	Publish time, ISO 8601 UTC.
durationSeconds	Video length in seconds.
views	Play count.
likes	Likes.
comments	Comments.
shares	Shares.
saves	Saves (favorites).
musicTitle	Sound used in the video.
coverUrl	Cover image URL.
transcript	Full transcript text.
wordCount	Words in the transcript.
language	Transcript language.
transcriptSource	tiktok_captions, speech_to_text, or none.
whisperModel	Whisper model used, when speech-to-text ran.
captionLanguages	Caption languages TikTok offers for the video.
segments	Timed lines: start and end in seconds, text.
srt	Transcript as SRT subtitles (when enabled).
note	Why a video has no transcript (no speech or no captions).
scrapedAt	When the row was scraped.

Export results as JSON, CSV, Excel, XML or HTML, or read them with the API.

Use cases

• Content research: find what top creators say in viral videos in your niche.
• AI and RAG pipelines: feed TikTok video text to LLMs for summaries, hooks and topic analysis.
• Brand monitoring: search transcripts for product and brand mentions.
• Repurposing: turn TikToks into blog posts, captions, threads or newsletters.
• Accessibility: generate SRT subtitles for videos.

How much does TikTok Transcript Scraper cost?

This Actor uses pay-per-event pricing: $0.002 per transcript ($2 per 1,000). Platform usage is included. When a video has no TikTok captions, Whisper speech-to-text adds $0.015 per started minute of audio (still about 3x cheaper than transcription-per-minute alternatives). A small run-start fee applies ($0.0015 per GB of memory). Videos without speech and failed videos are free.

Set Maximum cost per run in the run options to cap your spend; the Actor stops when the limit is reached.

Use TikTok Transcript Scraper with AI agents (MCP)

Give Claude, ChatGPT, Cursor or any MCP client this Actor as a tool through the hosted Apify MCP server:

$claude mcp add --transport http apify "https://mcp.apify.com/?tools=technicaldost/tiktok-transcript-scraper" --header "Authorization: Bearer YOUR_APIFY_TOKEN"

Or in a JSON MCP config:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com/?tools=technicaldost/tiktok-transcript-scraper",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

The tool is named technicaldost--tiktok-transcript-scraper. Ask your agent, for example, to "run TikTok Transcript Scraper with {"profiles": ["@nbcnews"], "maxVideosPerProfile": 2, "includeSrt": true}".

Run TikTok Transcript Scraper via API

curl -X POST "https://api.apify.com/v2/acts/technicaldost~tiktok-transcript-scraper/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"profiles": ["@nbcnews"], "maxVideosPerProfile": 2, "includeSrt": true}'

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("technicaldost/tiktok-transcript-scraper").call(run_input={"profiles": ["@nbcnews"], "maxVideosPerProfile": 2, "includeSrt": True})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('technicaldost/tiktok-transcript-scraper').call({"profiles": ["@nbcnews"], "maxVideosPerProfile": 2, "includeSrt": true});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Integrations

Schedule runs (daily, hourly) in Apify Console and send results to Google Sheets, Slack, Make, Zapier, n8n, Airbyte or your own webhook.

FAQ

Do I need a TikTok account or cookies?

No. The Actor reads public video pages as a logged-out visitor.

Where do the transcripts come from?

From TikTok's own captions when the video has them (field transcriptSource = tiktok_captions). Otherwise the audio is transcribed inside the Actor with OpenAI's open-source Whisper model (transcriptSource = speech_to_text). Nothing is sent to third-party AI services.

Which languages are supported?

TikTok captions come in the languages TikTok provides for the video. Whisper detects and transcribes 90+ languages automatically, or set Language to force one.

How many videos per creator?

Up to the creator's latest 10 public videos per username. For more, paste video URLs.

Which Whisper model should I use?

base (default) is fast and accurate for clear speech. small is more accurate for noisy audio or accents but about 3 times slower. tiny is the fastest.

Am I charged for videos with only music?

No. Videos without captions or speech are returned with a note and are free. Private, deleted or unreachable videos are listed in the ERRORS record and are free too.

Is it legal to scrape this data?

The Actor collects publicly available information only and does not log in or bypass access controls. Use the data in line with the website's terms and the laws that apply to you, especially when processing personal data.