YouTube Transcript & Subtitles API — Bulk Channels, Playlists, Search & SRT

Turn YouTube to text: extract transcripts (captions / subtitles) in bulk from single videos, whole channels, playlists and search results — as plain text, timestamped segments, SRT/VTT subtitle files or RAG-ready chunks. $3 per 1,000 transcripts, no start fee, and you only pay for successful transcripts.

Why this YouTube transcript API

	This Actor	Typical alternatives
Price per 1,000 transcripts	$3	$5–10
Start fee per run	None	Often charged
Failed videos (no captions, private, deleted)	Free	Often charged
Whole channels, playlists & YouTube search	Yes, with pagination	Often single videos only
Filter channels by Shorts / videos / lives and by date	Yes	Rare
Translation	YouTube captions free, AI translation optional	Rare
SRT / WebVTT subtitle files	Yes	Rare
RAG-ready chunks for LLMs	Yes	Rare
Video metadata (title, channel, date, views)	Included	Sometimes extra

What it does

• Accepts any YouTube URL format: watch?v=, youtu.be/, Shorts, /embed/, /live/, m.youtube.com, or bare 11-character video IDs.
• Takes channels (@handle, /channel/UC…, /c/…, /user/…) and playlists, and extracts the latest N videos (or all of them).
• Searches YouTube for keywords and extracts transcripts of the top results.
• Filters channels to regular videos, Shorts or past live streams only, and skips videos published before a date (skipped videos are free).
• Picks the best caption track for your preferred languages, preferring human-made captions over auto-generated ones.
• Optional translation to another language (see Translation).
• Removes duplicates, retries blocked requests with fresh residential IPs, and never fails the whole run because of a single video.

Input example

{
  "videos": ["https://www.youtube.com/watch?v=dQw4w9WgXcQ", "https://youtu.be/9Ff4S1FJdRM"],
  "channels": ["https://www.youtube.com/@veritasium"],
  "playlists": ["https://www.youtube.com/playlist?list=PLFs4vir_WsTwEd-nJgVJCZPNL3HALHHpF"],
  "searchQueries": ["black holes explained"],
  "maxVideosPerSource": 50,
  "channelContentType": "videos",
  "publishedAfter": "2026-01-01",
  "languages": ["en", "es"],
  "translateTo": "",
  "outputFormat": "both",
  "chunkSeconds": 60,
  "subtitleFormats": ["srt"]
}

Field	Default	Description
videos	–	Video URLs or IDs
channels	–	Channel URLs or handles
playlists	–	Playlist URLs or IDs
searchQueries	–	Keywords to search on YouTube (videos only)
maxVideosPerSource	50	Max videos per channel, playlist or search query (0 = all)
channelContentType	all	all, videos, shorts or streams (past live streams)
publishedAfter	–	Skip videos published before this date (YYYY-MM-DD); skipped videos are not charged
languages	["en"]	Preferred languages, in order. If none exists, the video's own language is returned
preferManual	true	Prefer human-made captions over auto-generated
translateTo	""	Language code to translate into (e.g. es)
aiTranslation	false	Translate with AI when YouTube can't (charged per 1,000 words, see Pricing)
outputFormat	both	text, segments or both
chunkSeconds	0	If > 0, adds chunks of ~N seconds for embeddings / RAG
subtitleFormats	–	srt and/or vtt: adds ready-to-use subtitle files
includeMetadata	true	Title, channel, publish date, duration, views, description, tags, thumbnail
maxConcurrency	10	Videos processed in parallel

Output example

One dataset item per video:

{
  "videoId": "abc123def45",
  "url": "https://www.youtube.com/watch?v=abc123def45",
  "status": "ok",
  "title": "How Black Holes Actually Work",
  "channelName": "Science Channel",
  "channelId": "UCxxxxxxxxxxxxxxxxxxxxxx",
  "publishedAt": "2026-05-01",
  "durationSeconds": 732,
  "viewCount": 1234567,
  "language": "en",
  "languageName": "English",
  "isAutoGenerated": false,
  "availableLanguages": ["en", "es", "pt-BR"],
  "wordCount": 1830,
  "text": "Imagine you could fall into a black hole. What would you actually see?…",
  "segments": [{ "start": 0.0, "duration": 3.2, "text": "Imagine you could fall into a black hole." }],
  "chunks": [{ "start": 0.0, "end": 60.4, "text": "Imagine you could fall into a black hole. What would you actually see?…" }]
}

status is one of:

• ok — transcript extracted (the only charged status)
• no_transcript — the video has no captions
• unavailable — private, deleted, age-restricted or region-blocked video (also used for channels/playlists that don't exist)
• error — invalid input or a failure after several retries; see the error field

A run summary (ok, noTranscript, unavailable, errors) is saved in the key-value store as SUMMARY.

Use cases

• RAG & chatbots: feed transcripts of a whole channel into a vector database using chunkSeconds.
• AI agents: give an agent the ability to "watch" YouTube videos through the transcript API.
• Content research & SEO: analyze what top creators talk about, find keywords and topics.
• Translated subtitles: get transcripts in your audience's language.
• Channel analysis: combine transcripts with views, dates and durations.

How to use it from code

API (cURL)

curl -X POST "https://api.apify.com/v2/acts/cazadores~youtube-transcript-bulk/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"videos": ["https://www.youtube.com/watch?v=dQw4w9WgXcQ"]}'

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")
run = client.actor("cazadores/youtube-transcript-bulk").call(run_input={
    "channels": ["https://www.youtube.com/@veritasium"],
    "maxVideosPerSource": 20,
    "outputFormat": "text",
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["title"], item.get("wordCount"))

JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_TOKEN' });
const run = await client.actor('cazadores/youtube-transcript-bulk').call({
    videos: ['https://youtu.be/dQw4w9WgXcQ'],
    chunkSeconds: 60,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items[0].chunks);

As an MCP tool for AI agents

Add the Apify MCP server to Claude, Cursor or any MCP client and enable this Actor as a tool:

https://mcp.apify.com/?actors=cazadores/youtube-transcript-bulk

Your agent can then call it with a list of video URLs and read the transcripts directly.

Pricing

Pay-per-event, no subscription and no start fee:

Event	Price
Successful transcript (status: ok)	$0.003 ($3 per 1,000) — lower on higher Apify plans
AI translation (only with aiTranslation on, and only when YouTube can't translate)	$0.003 per started 1,000 words
Videos without captions, unavailable videos, errors, videos skipped by date	Free
Listing channels, playlists & search, metadata, subtitle files, YouTube translation	Included

Proxy (residential) and compute costs are included in the price. Set a maximum cost per run in the run options and the Actor stops cleanly when it's reached.

Translation

When translateTo is set, the Actor:

1. Uses the video's own captions in that language when they exist (human-made or auto-generated).
2. Otherwise tries YouTube's automatic translation.
3. If YouTube doesn't provide a translation and aiTranslation is on, translates the transcript with AI line by line, so timestamps, chunks and SRT/VTT stay in sync with the video (translationSource: "ai").
4. Otherwise returns the transcript in its original language with a translationError field (still a successful transcript).

The fields translatedTo, translationSource (youtube_captions, youtube_auto_translate or ai) and originalLanguage tell you what happened for each video.

Known limitations

• Videos with no captions at all (common for Shorts, music and some live streams) return no_transcript. Speech-to-text for those videos is not included.
• Private, deleted, members-only, age-restricted and region-blocked videos return unavailable.
• YouTube's automatic translation is not available for every video; turn on aiTranslation to cover the rest (see Translation).
• Channel extraction uses the channel's uploads list, which includes regular videos, Shorts and past live streams.

Related actors

• TikTok Shop Scraper — bulk TikTok Shop product data: price, units sold, stock per variant, reviews and shop stats.

En español

Extractor de transcripciones de YouTube en lote: pegá URLs de videos, canales o playlists y obtené la transcripción (subtítulos) en texto, segmentos con tiempos o fragmentos listos para RAG. Cuesta $3 cada 1.000 transcripciones, sin costo de arranque, y solo pagás las transcripciones exitosas: los videos sin subtítulos, privados o borrados no se cobran. Incluye búsqueda por palabra clave, filtros por Shorts, videos o transmisiones y por fecha, archivos SRT/VTT, metadatos y traducción (gratis con los subtítulos de YouTube; con IA, opcional, a $0,003 cada 1.000 palabras).