YouTube Transcript API: Subtitles, SRT, Text & Markdown

$0.80 per 1,000 videos transcribed ($0.0008 per video). Pay per result, no monthly rental. Errors, blocked requests and videos without captions are free.

Extract YouTube transcripts (captions / subtitles) in bulk from video, Shorts, channel or playlist URLs. No API key, no login, no YouTube quota: a simple YouTube Data API alternative for transcripts. For every video you get:

• the timestamped transcript: start, duration and end in seconds + text for every caption line,
• the plain text transcript and an LLM-ready Markdown version with [mm:ss] timestamps,
• optional RAG chunks with timestamped YouTube links for citations,
• a ready-to-use SRT subtitle file with its download link,
• video metadata: title, channel, channel ID, duration, views, publish date, category, description, keywords, thumbnail,
• the list of available caption languages, and whether the transcript is auto-generated or translated.

Public data only: no account, no cookies, no browser. Proxy handling is automatic and included in the price.

How much does it cost?

You pay one video event per transcript delivered to the dataset:

Videos transcribed	Price
100	$0.08
1,000	$0.80
10,000	$8.00

Nothing is charged for videos without captions, private or removed videos, blocked requests, invalid URLs or videos skipped by onlyNew. Set a maximum cost per run in Apify: the actor stops cleanly when your budget is reached and never delivers anything it cannot bill.

How to use

1. Paste one or more URLs into YouTube URLs (videos, youtu.be links, Shorts, @handle channels, /channel/UC…, playlists, or bare video IDs).
2. Choose the transcript language (en, fr, de, es, pt, ja…).
3. Run. Results appear in the dataset (one item per video), SRT files in the key-value store.

Input example

{
  "urls": [
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
    "https://youtu.be/jNQXAC9IVRw",
    "https://www.youtube.com/@NASA",
    "https://www.youtube.com/playlist?list=PLxxxxxxxxxxxxxxxx"
  ],
  "language": "en",
  "fallbackToAnyLanguage": true,
  "translateIfMissing": false,
  "preferManualCaptions": true,
  "maxVideosPerChannelOrPlaylist": 10,
  "includeTimestamps": true,
  "includeMarkdown": true,
  "ragChunkWords": 300,
  "onlyNew": false,
  "saveSrtFile": true
}

Field	Default	Description
urls	–	Video, Shorts, channel or playlist URLs, or bare video IDs.
language	en	Preferred language code. en also matches en-US, en-GB.
fallbackToAnyLanguage	true	If your language is missing, return the original-language transcript.
translateIfMissing	false	If your language is missing, return YouTube's machine translation into it.
preferManualCaptions	true	Prefer creator-uploaded subtitles over auto-generated ones.
maxVideosPerChannelOrPlaylist	20	Videos per channel (latest uploads) or playlist.
includeTimestamps	true	Include the segments array.
includeMarkdown	true	Include the LLM-ready markdown field.
ragChunkWords	0	If > 0, include chunks of about this many words (for embeddings / RAG).
onlyNew	false	Skip videos already delivered by your previous runs (scheduled runs).
saveSrtFile	true	Save one .srt file per video and return its link.
fetchPublishDate	true	Add publishDate, uploadDate, category.
maxConcurrency	5	Videos processed in parallel.
residentialFallback	true	Retry blocked videos through residential proxy (no extra charge).

Output example

{
  "status": "success",
  "videoId": "jNQXAC9IVRw",
  "url": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
  "title": "Me at the zoo",
  "channelName": "jawed",
  "channelId": "UC4QobU6STFB0P71PMvOGN5A",
  "channelUrl": "https://www.youtube.com/channel/UC4QobU6STFB0P71PMvOGN5A",
  "durationSeconds": 19,
  "viewCount": 350000000,
  "publishDate": "2005-04-23",
  "uploadDate": "2005-04-23",
  "category": "Film & Animation",
  "description": "The first video on YouTube...",
  "keywords": [],
  "thumbnailUrl": "https://i.ytimg.com/vi/jNQXAC9IVRw/hqdefault.jpg",
  "isLiveContent": false,
  "language": "en",
  "languageName": "English",
  "isAutoGenerated": false,
  "isTranslated": false,
  "availableLanguages": [
    { "code": "en", "name": "English", "isAutoGenerated": false },
    { "code": "de", "name": "German", "isAutoGenerated": false }
  ],
  "segmentCount": 6,
  "wordCount": 48,
  "text": "All right, so here we are in front of the elephants...",
  "segments": [
    { "start": 1.2, "duration": 2.16, "end": 3.36, "text": "All right, so here we are in front of the elephants" }
  ],
  "markdown": "# Me at the zoo\n\n- Channel: jawed\n- Published: 2005-04-23\n- Duration: 00:19\n- Language: en\n- URL: https://www.youtube.com/watch?v=jNQXAC9IVRw\n\n## Transcript\n\n**[00:01]** All right, so here we are in front of the elephants...\n",
  "chunks": [
    { "index": 0, "start": 1.2, "end": 17.5, "wordCount": 48, "url": "https://youtu.be/jNQXAC9IVRw?t=1", "text": "All right, so here we are..." }
  ],
  "srtFileUrl": "https://api.apify.com/v2/key-value-stores/<storeId>/records/jNQXAC9IVRw-en.srt",
  "srtFileKey": "jNQXAC9IVRw-en.srt",
  "sourceUrl": "https://youtu.be/jNQXAC9IVRw",
  "scrapedAt": "2026-10-15T09:00:00+00:00"
}

Videos that cannot be transcribed still get a free item so you know why:

{
  "status": "error",
  "videoId": "abcdefghijk",
  "url": "https://www.youtube.com/watch?v=abcdefghijk",
  "sourceUrl": "https://www.youtube.com/@somechannel",
  "error": "this video has no captions/subtitles"
}

Filter on status = "success" to keep billed results only. The Overview view shows a table with thumbnail, title, channel, language, transcript and SRT link.

Use with AI agents (MCP)

This actor is MCP-ready: Claude, Cursor, ChatGPT and other AI agents can call it through the Apify MCP server (https://mcp.apify.com). The input is just a list of URLs plus a language code; the output is flat JSON with text, markdown (ready to drop into a prompt) and optional chunks with timestamped links for grounded answers. Typical agent prompts: "summarize the last 10 videos of @channel", "find where this video talks about pricing and give me the timestamp".

Integrations

• API: run it from any language with the Apify API or the apify-client for Python / JavaScript and read the dataset as JSON, CSV, Excel or XML.
• Make, Zapier, n8n: use the Apify app/node to start a run and get the transcripts in your workflow.
• Google Sheets: export the dataset or use the Apify integration to append one row per video.
• Schedules + onlyNew: schedule a channel or playlist daily and receive only the new uploads, billed only once.
• Webhooks: get notified when a run finishes.

Use cases

• AI & LLM pipelines: feed transcripts to summarizers, chatbots, RAG knowledge bases or AI agents.
• Content repurposing: turn videos into blog posts, newsletters, social threads or show notes.
• Subtitles: download SRT files for editing, translation or re-upload.
• Research & monitoring: track what a channel talks about, follow keywords and brand mentions over time.
• SEO: publish searchable text versions of your own videos.
• Education: build study notes and searchable lecture archives from playlists.

FAQ

Which videos work? Any public video that has captions (creator-made or YouTube auto-generated), including Shorts and finished live streams.

What am I charged for? Exactly one video event ($0.0008) per transcript delivered to the dataset. Errors, videos without captions, blocked requests and duplicates are free.

Do I need a YouTube API key, an account or a proxy? No. No key, no login, no quota. Proxy handling (datacenter first, residential fallback when YouTube blocks) is automatic and included in the price.

What if my language is not available? By default you get the original-language transcript (the language field tells you which). Enable translateIfMissing to get YouTube's machine translation, or disable fallbackToAnyLanguage to get an error instead.

Auto-generated or human captions? Human captions are preferred by default; isAutoGenerated tells you which one you got.

How do channels work? A channel URL returns its latest uploads (videos, Shorts and lives), up to maxVideosPerChannelOrPlaylist.

How does onlyNew work? Delivered video IDs are remembered per language in your key-value store named youtube-transcripts-state. Later runs skip them (for free). Delete that store to start over.

Does it transcribe videos without captions (speech-to-text)? No. It returns the captions YouTube already has; videos without any are reported as free errors.

Limitations

• Private, members-only and age-restricted videos require sign-in and are not supported (returned as free errors).
• Videos with no captions at all cannot be transcribed (no speech-to-text).
• Live streams in progress usually have no transcript until they end.
• Channel listing follows the uploads playlist order (newest first); large channels are limited by maxVideosPerChannelOrPlaylist.
• YouTube sometimes rate-limits cloud servers; the actor retries with new IPs and a residential fallback. Very long videos (3 h+) that end up on the residential fallback may fail (free).