YouTube Transcript Scraper

Turn any list of YouTube videos into clean transcripts — plain text and timestamped segments — in one run. Pick the caption language, let YouTube auto-translate, process hundreds of videos at once, and pay only for the transcripts you actually receive. Built for AI summaries, RAG pipelines, content repurposing, subtitles and research.

What does the YouTube Transcript Scraper do?

Paste YouTube URLs or video IDs and the actor fetches each video's captions — human-made or auto-generated — straight from YouTube, without downloading video or audio. For every video you get the full transcript as text, the same transcript as segments with start times and durations, the caption language, whether it was auto-generated, the list of all available caption languages, and the video's title, channel and duration.

• 🎯 Bulk — hundreds of videos per run, processed in parallel (maxConcurrency).
• 🌍 Language control — preferred languages in priority order (en, en-US, es, …), human-made captions preferred over auto-generated, optional fallback to any available language.
• 🔁 Translation — ask for YouTube's automatic translation of the chosen track (translateTo: "fr").
• ⏱ Timestamps — segments with start, duration and text, ready for subtitle files, clip search or chaptering.
• 💸 Fair pricing — videos without captions (or private/unavailable ones) are returned as found: false rows and are not charged.
• 🤖 Agent-ready — JSON output with a documented schema, callable through the Apify API, Make, Zapier, n8n or an AI agent via Apify's MCP server.

What data does the scraper return?

Field	Description
videoId, url	Video ID and canonical watch URL
title, channel, channelId, durationSeconds	Video metadata
found	true when a transcript was delivered
language, languageName, isGenerated	Caption track used and whether it is auto-generated
translatedTo	Language code when translateTo was applied, else null
transcript	The full transcript as one plain-text string
wordCount, segmentCount	Size of the transcript
segments	Array of { "start", "duration", "text" } (seconds), when includeSegments is on
availableLanguages	Every caption track YouTube offers for the video: { "code", "name", "isGenerated" }
warning	Caveat on a delivered transcript, e.g. the requested translation was refused by YouTube and the original language was returned (null otherwise)
error	Why no transcript was delivered (null on success)

How to get a YouTube transcript — step by step

1. Click Try for free (a free Apify account takes a minute).
2. Paste your video URLs or IDs into YouTube video URLs or IDs, one per line. Watch URLs, youtu.be links, Shorts, embeds and bare IDs all work.
3. Optionally set Preferred caption languages (default en) and Translate to.
4. Click Start. Results appear in the Dataset tab; export them as JSON, CSV, Excel or XML, or fetch them via the API.

How much does it cost to scrape YouTube transcripts?

$4 per 1,000 transcripts ($0.004 each), charged per transcript delivered. Videos with no captions, private or removed videos cost nothing. Apify's free plan includes $5 of monthly usage, enough for roughly 1,000 transcripts per month at no cost — no credit card needed to try.

Input example

{
  "videoUrls": [
    "https://www.youtube.com/watch?v=aircAruvnKk",
    "https://youtu.be/jNQXAC9IVRw",
    "dQw4w9WgXcQ"
  ],
  "languages": ["en"],
  "translateTo": "",
  "includeSegments": true
}

Output example

{
  "videoId": "aircAruvnKk",
  "url": "https://www.youtube.com/watch?v=aircAruvnKk",
  "title": "But what is a neural network? | Deep learning chapter 1",
  "channel": "3Blue1Brown",
  "channelId": "UCYO_jab_esuFRV4b17AJtAw",
  "durationSeconds": 1120,
  "found": true,
  "language": "en",
  "languageName": "English",
  "isGenerated": false,
  "translatedTo": null,
  "transcript": "This is a 3. It's sloppily written and rendered at an extremely low resolution of 28 by 28 pixels …",
  "wordCount": 3357,
  "segmentCount": 286,
  "segments": [
    { "start": 4.22, "duration": 1.18, "text": "This is a 3." },
    { "start": 5.4, "duration": 2.3, "text": "It's sloppily written and rendered at an extremely low resolution of 28 by 28 pixels" }
  ],
  "availableLanguages": [
    { "code": "en", "name": "English", "isGenerated": false },
    { "code": "es", "name": "Spanish", "isGenerated": false }
  ],
  "error": null
}

A video without captions is returned as { "videoId": "…", "found": false, "error": "No captions available for this video", … } — and is not charged.

What can you build with YouTube transcripts?

• AI summaries and chat with videos — feed transcripts to Claude, GPT or an open model; segments let the model cite timestamps.
• RAG and knowledge bases — index channels, courses and conference talks for retrieval.
• Content repurposing — turn videos into blog posts, newsletters, social posts or SEO articles.
• Subtitles and accessibility — generate .srt/.vtt files from segments.
• Research and monitoring — search what is said across many videos, track mentions of brands, products or topics.
• Language learning — original captions plus translateTo for parallel texts.

Use it as a tool for AI agents

Through Apify's MCP server, agents can call this actor with a list of videos and receive transcripts in a stable JSON schema — no browser automation, no audio transcription delay, and the found/error fields make failures explicit instead of hallucinated.

FAQ

Does it work with auto-generated captions?

Yes. If a video only has YouTube's automatic captions, you get those and the row says isGenerated: true. When both exist, human-made captions win by default (preferManual).

Which languages are supported?

Every caption track YouTube offers for the video — availableLanguages lists them. Use languages to set your preference order; translateTo returns YouTube's automatic translation for translatable tracks. YouTube rate-limits translations much harder than plain captions: if it refuses one, you still get the original-language transcript, with the reason in warning, and you are charged once as usual.

Why is a transcript missing for some videos?

The video has no captions at all, is private, removed or age-restricted, or YouTube temporarily refused the request. The error field says which. Missing transcripts are never charged. Playlist and channel URLs are not supported yet — list the video URLs (our Google Search scraper with site:youtube.com queries is a quick way to collect them).

Does it download the video or audio?

No. Only the caption data is fetched, so runs finish in seconds and no copyrighted media is copied. For videos without captions, pair it with a speech-to-text service.

Is scraping YouTube transcripts legal?

Captions are publicly available data, and this actor only reads what YouTube serves to any viewer. You are responsible for how you use the content — respect copyright and YouTube's terms of service, and do not republish full transcripts you do not have rights to.

Can I get SRT or VTT subtitles?

Yes — export the dataset as JSON and convert segments (start and duration in seconds) to SRT/VTT; a few lines of code or any subtitle tool will do.

More data tools by damilo

• Google Search Results Scraper — find videos, articles and competitors for any keyword (site:youtube.com queries work great with this actor).
• Google Autocomplete Scraper — keyword ideas and search intent for your video and content research.
• Google Images Scraper — image results with full-size URLs for thumbnails and visual research.
• Google Maps Scraper — local business data for lead generation.
• Google Shopping Scraper, Amazon Search Scraper, eBay Search Scraper — product and price data.
• Y Combinator Scraper — startups, founders and jobs.
• FDA Acronyms & Abbreviations Database — free regulatory acronym lookup.

Support

Questions, missing features (playlists, channels, SRT export) or a video that should have worked? Use the Issues tab — it's answered quickly. If the actor saved you time, a ⭐⭐⭐⭐⭐ review helps other people find it.