One run for all three platforms: YouTube Shorts, TikTok and Instagram Reels turned into text — transcript with timestamps, on-screen captions and the first-3-second hook. Shorts use YouTube's own captions, so they cost less. Paste links or whole profiles; no login.
transcript | Spoken transcript |
segments | Timestamped segments |
hook | What is said in the first 3 seconds |
onScreenText | Text burned into the video |
author | Video author |
views / likes / comments | Engagement counts |
caption | Post caption |
publishedAt | Publish date |
authorComparison | How a video compares with the author's usual |
$1.00 per 1,000 rows
Pay per event: you are charged only for what the run returns — videos processed, 10-second blocks of recognised audio, on-screen text reads. There is no monthly subscription; caps in the input keep each run's cost predictable.
Apify gives $5 free credits every month — enough to test this scraper at no cost.
Ready-made inputs. Copy one into the scraper form or send it through the API.
Uses TikTok's own captions first, with segments.
{
"videoUrls": [
"https://www.tiktok.com/@duolingo/video/7683994892506270989"
],
"transcriptionMode": "captions-first",
"includeTranscriptSegments": true
}Latest 20 videos compared with the author's usual.
{
"profiles": [
"@duolingo"
],
"resultsLimitPerProfile": 20,
"compareWithAuthor": true
}Spoken transcript plus text burned into the picture.
{
"profiles": [
"@duolingo"
],
"resultsLimitPerProfile": 10,
"includeOnScreenText": true,
"ocrFps": "1.5",
"language": "en"
}Videos since September 2026, up to 2 minutes long.
{
"profiles": [
"@duolingo"
],
"onlyPostsNewerThan": "2026-09-01",
"transcriptionMode": "ai-only",
"quality": "accurate",
"maximumVideoDurationSeconds": 120
}TikTok, Instagram Reels and YouTube Shorts, by video link or whole profile/channel.
The words spoken in the first three seconds of the video, returned as a separate field.
The scraper reads publicly available TikTok, Instagram and YouTube data without logging in. Scraping public data is generally allowed, but you are responsible for complying with local laws and the site's terms, and for not collecting or misusing personal data.
Text burned into the picture — caption cards, title cards and hard subtitles — read separately from what the author says.
Call it through the Apify API, use the Python or JavaScript client, or connect it to Make, n8n, Zapier or Google Sheets. Results export to JSON, CSV or Excel, and runs can be scheduled.