TikTok Transcript Scraper: Video to Text with Timestamps

What does TikTok Transcript Scraper do?

TikTok Transcript Scraper turns TikTok videos into clean text transcripts with timestamps. Paste video links, or a creator's profile to get their latest videos, and get one row per video with the full transcript, timestamped segments, the language, the author, caption, duration and engagement stats.

Most TikToks carry TikTok's own captions (creator-written or TikTok's speech recognition). Those are read directly, which is fast and cheap. Videos without captions are transcribed from their audio with Whisper.

Run it in Apify Console, call it from the API, schedule it, or connect it to Make, Zapier, n8n, Google Sheets or your own code.

What data can you extract from a TikTok video?

📝 Full transcript	⏱️ Timestamped segments	🗣️ Language
👤 Author	📰 Title and caption	⏳ Duration
🔢 Word count	🎙️ Captions or speech-to-text	👀 Views
❤️ Likes	💬 Comments	🔁 Shares
📅 Published at	🖼️ Thumbnail	🔗 Video URL

Why use this TikTok Transcript Scraper to download transcripts?

• Turn TikToks into blog posts, scripts and captions.
• Study hooks and scripts of viral videos.
• Feed creator videos into an AI knowledge base.
• Monitor spoken brand mentions.
• Pay only for transcripts. Private, deleted and silent videos are free.

How to get TikTok transcripts (step by step)

1. Click Try for free and open the Input tab.
2. Paste video URLs or profiles.
3. Click Start. A few videos take seconds.
4. Download the results as JSON, CSV, Excel or HTML, or read them through the API.

Input

{
    "urls": [
        "https://www.tiktok.com/@tiktok/video/7106594312292453675"
    ],
    "profiles": [],
    "maxVideosPerProfile": 10,
    "speechToText": "fallback",
    "language": "",
    "includeSegments": true
}

All fields are described on the Input tab. Only TikTok videos are accepted; links from other platforms come back free with a pointer to the matching actor.

Output

One row per video. Download the dataset as JSON, HTML, CSV or Excel.

Field	Description
url, videoId	The video.
author, authorName, authorUrl	Creator handle, display name and profile link.
title, caption	Title or post caption.
durationSec, publishedAt, thumbnailUrl	Length, publish date, cover image.
transcript	The full transcript as one text.
segments	{ start, end, text } with times in seconds.
language	Two-letter language code of the transcript.
transcriptSource	subtitles (platform captions) or speech-to-text (Whisper).
stats	Views, likes, comments and shares when the platform shows them.
status, errorCode, message	success, no_transcript or failed, with a reason such as private_or_removed or no_speech.
charged	Whether this row was billed.

How much does it cost to get TikTok transcripts?

Pay per transcript, no subscription:

• $0.90 per 1,000 transcripts read from captions (most videos).
• $2.50 per 1,000 speech-to-text transcripts (videos without captions), first minute of audio included. Longer videos add $0.0025 per extra started minute.
• A $0.005 start fee per run.

Failed, private, removed and silent videos are free. Higher Apify plans get up to 20% off.

Subtitle files, filters and budget

• SRT and WebVTT: pick them under Subtitle files to get ready-to-use subtitle text in the srt and vtt fields, for video editors, players and translation tools.
• Profile filters: Only videos from the last N days and Minimum views skip older or smaller videos before anything is charged.
• Speech-to-text budget: Speech-to-text budget (minutes) caps how much audio one run transcribes, so a long video list can never surprise you.
• AI fallback: videos with captions use them; only videos without captions go to speech-to-text.

Run it on a schedule

Save a task with the creators you follow, set Only videos from the last N days to 1 or 7, and add a daily or weekly schedule: you get every new video transcribed automatically. Connect Google Sheets, Slack or a webhook to receive the transcripts as soon as the run finishes.

Use it with AI tools, n8n and MCP

• RAG and LangChain: load the dataset with Apify's LangChain or LlamaIndex loaders and index the transcript field.
• n8n, Make and Zapier: use the Apify node to run the actor and send each transcript to Google Sheets, Notion or Slack.
• MCP: add the Apify MCP server (https://mcp.apify.com) to Claude, ChatGPT or Cursor and ask for transcripts in plain language.

FAQ, limits and support

Which videos can't be transcribed? Private and deleted videos cannot be read. Photo slideshows and music-only videos come back free (no_video, no_speech).

How accurate is it? Speech-to-text uses Whisper large-v3 turbo, which is very accurate on clear speech in 90+ languages. Music and heavy background noise lower accuracy.

Is there an official TikTok API for transcripts? How is this different? TikTok has no public API that returns the spoken text of any public video. This actor gives you the transcript, timestamps and video stats from a URL or a profile, with no developer account and no app review.

What are the alternatives to TikTok Transcript Scraper? Copying captions by hand, uploading videos to a transcription service one by one, or other transcript actors on Apify. This one works in bulk, reads captions first when they exist, and charges nothing for failed or silent videos.

Is it legal? The actor only reads public videos that anyone can watch without logging in. Respect copyright, platform terms and privacy laws such as the GDPR.

Found a problem or need a field? Open an Issue on the actor page.

Other transcript scrapers

Actor	Platform
Instagram Transcript Scraper	Instagram Reels and video posts
YouTube Transcript Scraper	YouTube videos and Shorts
X (Twitter) Video Transcript Scraper	X (Twitter) videos
Facebook Video Transcript Scraper	Facebook videos and Reels
Douyin Transcript Scraper	Douyin (抖音) videos
Video Transcript Scraper	TikTok, Instagram, YouTube and X in one actor

Privacy

This Actor sends anonymous run inputs (such as video URLs) and a hashed account ID to the developer to see which inputs are useful. Tokens, cookies and proxy settings are never sent. Nothing from your results is collected.