YouTube Transcript Extractor — RAG & AI Ready

Extract clean, structured transcripts from any YouTube video in seconds. No YouTube API key required. Batch-ready. Built for AI pipelines that need to index video content at scale.

---

Why developers choose this actor

Feature	This actor
API key required	❌ None needed
Batch processing	✅ Hundreds of videos per run
Parallel fetching	✅ Up to 10 concurrent videos
Timestamp support	✅ Optional [MM:SS] per line
Multi-language	✅ Any language, priority order
Auto-retry on block	✅ 3 attempts, fresh proxy each time
LLM-ready output	✅ Clean paragraphs, no post-processing
Cloud IP bypass	✅ Residential proxy routing built-in

---

Input

Field	Type	Required	Description
videoUrl	string	one of the two	Single YouTube URL or video ID
videoUrls	array	one of the two	List of URLs/IDs for batch mode
includeTimestamps	boolean	no	Prefix each line with [MM:SS] (default: false)
languages	array	no	Language priority list (default: ["en","en-US","en-GB"])
maxConcurrency	integer	no	Parallel videos, 1–10 (default: 3)
proxyConfiguration	object	recommended	Use Apify Residential proxies to bypass YouTube cloud IP blocks

Supported URL formats:

https://www.youtube.com/watch?v=VIDEO_ID
https://youtu.be/VIDEO_ID
https://www.youtube.com/shorts/VIDEO_ID
https://www.youtube.com/embed/VIDEO_ID
VIDEO_ID                                   ← bare 11-character ID also works

---

Output

Each video produces one result object.

Successful transcript:

{
  "videoId": "jNQXAC9IVRw",
  "youtubeUrl": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
  "language": "en",
  "wordCount": 39,
  "durationMinutes": 0.3,
  "segmentCount": 9,
  "transcript": "All right, so here we are in front of the elephants...",
  "status": "success"
}

Failed video (disabled captions, private, etc.):

{
  "videoId": "abc123",
  "youtubeUrl": "https://www.youtube.com/watch?v=abc123",
  "status": "error",
  "error": "Transcripts are disabled for this video"
}

Field	Description
videoId	11-character YouTube video ID
youtubeUrl	Full YouTube URL
language	Language code of the fetched transcript
wordCount	Total word count of the transcript
durationMinutes	Video duration in minutes
segmentCount	Number of caption segments
transcript	Clean Markdown text, ready for LLM ingestion
status	success or error
error	Error description (only present on failure)

---

Example inputs

Single video:

{
  "videoUrl": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Batch of videos with timestamps:

{
  "videoUrls": [
    "https://www.youtube.com/watch?v=VIDEO_1",
    "https://www.youtube.com/watch?v=VIDEO_2",
    "https://www.youtube.com/watch?v=VIDEO_3"
  ],
  "includeTimestamps": true,
  "maxConcurrency": 5,
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Non-English content:

{
  "videoUrl": "https://www.youtube.com/watch?v=VIDEO_ID",
  "languages": ["fr", "fr-FR", "en"],
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

---

Use cases

• RAG knowledge bases — ingest entire YouTube channels into vector databases (LangChain, LlamaIndex, Pinecone, Weaviate)
• AI video summarization — feed transcripts to GPT-4 / Claude for summaries, key points, action items
• Content repurposing pipelines — convert video content to blog posts, newsletters, social media threads
• Podcast transcription — extract transcripts from YouTube-hosted podcast episodes
• Competitive intelligence — monitor competitor product demos, webinars, conference talks
• Educational tools — index courses and lectures for search and Q&A
• Multilingual pipelines — extract captions in the original language for translation workflows

---

Code examples

Python (Apify client)

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")

run = client.actor("foudhilriahi/youtube-transcript-extractor").call(
    run_input={
        "videoUrls": [
            "https://www.youtube.com/watch?v=VIDEO_1",
            "https://www.youtube.com/watch?v=VIDEO_2",
        ],
        "maxConcurrency": 5,
        "proxyConfiguration": {
            "useApifyProxy": True,
            "apifyProxyGroups": ["RESIDENTIAL"],
        },
    }
)

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    if item["status"] == "success":
        print(f"{item['videoId']}: {item['wordCount']} words")
        print(item["transcript"][:200])

LangChain integration

from langchain_community.utilities import ApifyWrapper

apify = ApifyWrapper()

loader = apify.call_actor(
    actor_id="foudhilriahi/youtube-transcript-extractor",
    run_input={
        "videoUrls": ["https://www.youtube.com/watch?v=VIDEO_ID"],
        "proxyConfiguration": {
            "useApifyProxy": True,
            "apifyProxyGroups": ["RESIDENTIAL"],
        },
    },
    dataset_mapping_function=lambda item: item.get("transcript", ""),
)

docs = loader.load()
# docs is now ready for your RAG pipeline

n8n / Make automation

1. Add an Apify node
2. Actor ID: foudhilriahi/youtube-transcript-extractor
3. Input: { "videoUrl": "{{ $json.youtubeUrl }}", "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] } }
4. Connect output to your vector database, Google Sheets, or email node
5. Schedule daily — done

---

Pricing

Pay Per Event — you only pay for successful transcripts.

Volume	Price per transcript
Any	$0.05 per successful extraction

Failed videos (disabled captions, private videos, no captions available) are not charged.

Cost example: 1,000 transcripts = $50. At a typical video length of 30 minutes, that's 500,000 minutes of transcribed content for $50.

---

Notes & limitations

• Videos must have captions available (manually added or auto-generated by YouTube)
• Private and age-restricted videos cannot be transcribed
• YouTube blocks cloud IP ranges — residential proxy configuration is required for reliable operation (pre-filled in the input)
• Auto-generated captions may contain minor transcription errors, especially for technical terms

---

Keywords

youtube transcript extractor, youtube transcript api, youtube to text, youtube captions download, video transcript, rag youtube, llm video content, youtube to markdown, ai video pipeline, langchain youtube, llamaindex video, youtube transcript python, video indexing ai, batch youtube transcript, youtube captions api