image
Coin

1

free
new

YouTube Transcript Scraper

Get the full transcript of any YouTube video — every caption language, playlists & channels too.

Download sample

Trusted worldwide by the best

Export transcripts from YouTube — right now

Extract

Paste any YouTube video, playlist or channel URL and get the full transcript — every caption line with its start and end time, in any language the video offers — plus title, description, views, likes, duration and channel details: 34 data points per video.

image

Schedule

Automate YouTube data collection on daily, weekly, or monthly schedules.

image

Export

Export all transcript data to Google Sheets, email, or S3 — fully automated.

image

Built for Agents

Run it from your stack.
Or hand it to your agent.

YouTube Transcript Scraper ships with a Python SDK, a one-line CLI and a native MCP server your agents can plug into directly — no rate limits to dodge, no bot mitigation to bypass, no parsing errors to chase down.

Read the docs
Get your API key
# install once (Python 3.10+)
pip install lobstrio

# check your token
lobstr whoami

# find the scraper
lobstr crawlers search "YouTube"

# one call — launch, wait, download
lobstr go youtube-transcript-scraper \
  "<any YouTube URL>" \
  --param max_results=200 \
  -o results.csv

34 data attributes per transcript

Every scrape returns structured data you can export as CSV, JSON, or via API.

Video Id

YouTube video identifier

iG9CE55wbtY
Export Transcripts
url

Canonical URL of the video

https://www.youtube.com/watch?v=iG9CE55wbtY
Export Transcripts
title

Video title

Do schools kill creativity? | Sir Ken Robinson | TED
Export Transcripts
description

Full video description

Visit http://TED.com to get our entire library of TED Talks...
Export Transcripts
keywords

Keywords (tags) set by the creator

[
  "Ken",
  "Robinson",
  "TED",
  "creativity"
]
Export Transcripts
Duration Seconds

Video length in seconds

1,203
Export Transcripts
View Count

Exact view count at collection time

24949463
Export Transcripts
Like Count

Exact like count at collection time (null when likes are hidden)

306922
Export Transcripts
Comment Count Text

Approximate comment count as displayed by YouTube (YouTube exposes no exact number)

13K
Export Transcripts
Published At

Publish date and time (UTC)

Jan 7, 2007
Export Transcripts
image

Fast

12 transcripts per minute
Lightning-fast performance

image

Solid

99.95% task success rate
Dependable, reliable data every time

image

Cost-competitive

$0.5 per 1000 transcripts
Top affordability worldwide

Live

Built to run.

Most scrapers break in silence. This one doesn't — and we show it, openly.

100.00%

incident-free runs · last 90 days

0
incidents · 90 days
100%
resolved
—
median time to fix
90 days agoToday
Operational
Degraded
Down
Resolved incident
Open incident
How we measure →

Tracked from every run. Incidents stay on the record even after they're fixed.

Simulator plan

Trusted by the best

Used obsessively by the data-hungry around the world

★★★★★
★★★★★
★★★★★
★★★★★
★★★★★
★★★★★

Frequently Asked Questions

Is it legal to scrape YouTube transcripts?

Yes — public captions are public data. A transcript is text YouTube already serves to anyone who taps the caption button, and nothing private or unlisted is touched. The detail is in our guide to the legality of web scraping.

Do I need a YouTube account or login?

No login required. No account, no cookie, no Chrome extension and no API key of your own to look after. Paste a video, playlist or channel URL and the text starts landing in your spreadsheet.

What data do I get for each video?

The full transcript plus 34 data points. Every caption line with its start and end time, the word count, the language used and the ones on offer — then title, description, keywords, duration, views, likes, comments, publish date and the channel behind it.

How fast is it?

About 12 transcripts a minute. One video is back in seconds, and a hundred-video channel is done in roughly the time it takes to watch one of them. Each transcript costs 1 credit, however long the video runs.

Can I scrape a whole channel or playlist at once?

Yes — paste a playlist or channel URL and it opens out into the videos. One row per video, @handles and channel links alike. Max Results decides how many you take: 50 by default, or leave it empty and take the lot.

Which caption languages can I download?

Any language the video offers. Stay on the original track, ask for every track at once, or name a single language and get only that one — a video without it comes back flagged, never silently empty. Each row lists what was available.

Are auto-generated captions included?

Yes, switched on by default. YouTube's own speech-to-text tracks come along with the captions creators uploaded, which is how you get usable text out of videos nobody ever subtitled. Turn it off for hand-written captions only.

Where does the data land once the run is done?

Pretty much anywhere you already work. Google Sheets, Make, n8n, Amazon S3, email, webhooks — plus our MCP server, so Claude or ChatGPT can launch a run and read the transcripts itself. CSV, Excel and JSON downloads too. See all integrations.

Ready to get started?

Export your first transcripts for free.

Contact sales