Video Transcript Scraper – YouTube, X, TikTok, Facebook

Video Transcript Scraper — full text, segments and video data

Video Transcript Scraper takes video, playlist or channel links from YouTube, X, TikTok, Facebook, Vimeo, Dailymotion, Loom, Rumble and similar sites, and returns each video's transcript as plain text and as timestamped segments, together with the title, channel, views, likes, languages and more. You can mix platforms in one run and paste as many links as you like. Every response is structured JSON — ready to pass directly to an LLM, index into a vector store, or feed a RAG pipeline. No account, login or cookies are needed, because it reads the captions that videos already publish. Use it to turn video libraries into searchable text.

Why choose this Video Transcript Scraper?

• Bulk and mixed links. One run can take videos, playlists, channels and @handles from different platforms.
• Real-time output. Every finished video appears in the output table straight away, so you can watch the table fill while the run goes on.
• Separate tables per section. Besides the transcript table you get tables for timestamped segments, caption languages, channels, your links and videos without a transcript.
• Starts light, escalates only if needed. The run begins without a proxy. If a platform refuses the request, it switches to a datacenter proxy, then to a residential proxy, and keeps that proxy for the rest of the run.
• Honest about gaps. Videos with no captions are never invented; they are listed with the reason, or kept with empty text if you turn on best effort.

Key features

• Full transcript as one string (text) plus timestamped transcript segments with start, end and a jump link to that moment (YouTube and Vimeo).
• Choose a caption language; if it is not available the closest one is returned and language_note says so.
• Manual (human-written) captions are preferred over auto-generated ones, and transcript_source tells you which you got.
• Playlist and channel expansion with a total video limit.
• Duplicate links are removed, and youtu.be/ID and watch?v=ID count as the same video.
• Plain-language reasons for failures: private, removed, age-restricted, login required, region-locked, live, no captions, blocked.

Supported platforms

Platform	What to expect
YouTube	Videos, Shorts, playlists, channels and @handles; manual and auto captions
Dailymotion	Videos that carry captions
X, TikTok, Loom	Video details are returned; a transcript is returned when the video carries captions
Facebook, Vimeo, Rumble	Accepted, but these sites often refuse automated requests or ask for a login, so some videos come back as failed

Only captions that already exist are used. The Actor does not transcribe audio.

Input

Example input:

{
    "urls": [
        "https://www.youtube.com/watch?v=vl6jn-DdafM",
        "https://www.youtube.com/@NASA/videos"
    ],
    "maxVideos": 10,
    "language": "auto",
    "bestEffort": false,
    "proxyConfiguration": { "useApifyProxy": false }
}

Field	Type	What it does
🎬 urls	array, required	Video, playlist or channel links, one per line. Platforms can be mixed.
📊 maxVideos	integer, default 10	Total videos for the whole run. 0 means every video found. Free plan: up to 10 videos per run.
🌍 language	string, default auto	Caption language code such as en, es, fr. auto returns the video's own language.
🧩 bestEffort	boolean, default false	Keep videos without captions in the main table, with details and an empty transcript.
🌐 proxyConfiguration	object	Optional. Default is no proxy; the run escalates by itself if the platform refuses.

Output

Each video is one row in the Transcripts table. Example (shortened):

{
    "url": "https://www.youtube.com/watch?v=vl6jn-DdafM",
    "status": "ok",
    "platform": "youtube",
    "video_id": "vl6jn-DdafM",
    "title": "We Are Going",
    "channel": "NASA",
    "channel_id": "UCLA_DiR1FfKNvjuUpBHmylQ",
    "channel_followers": 15100000,
    "published_date": "2019-05-14T20:34:54Z",
    "duration": 227,
    "duration_string": "3:47",
    "view_count": 17934893,
    "like_count": 1337804,
    "selected_language": "en",
    "transcript_source": "manual",
    "word_count": 591,
    "words_per_minute": 156,
    "text": "- [Narrator] 50 years ago we pioneered a path to the moon. The trail we blazed cut through ...",
    "transcript": [
        { "start": 2.36, "end": 5.26, "text": "- [Narrator] 50 years ago we pioneered a path to the moon.", "link": "https://www.youtube.com/watch?v=vl6jn-DdafM&t=2s" }
    ]
}

Table	One row per	Main fields
Transcripts (default)	video	title, channel, view_count, like_count, selected_language, transcript_source, word_count, text, transcript
Transcript segments	caption segment	segment_number, timestamp, start, end, text, link
Caption languages	language a video offers	language, type (manual or auto), selected
Channels	channel or uploader	channel, channel_url, channel_followers
Your links	link you gave	kind, videos_found, status, note
Videos without a transcript	video that gave none	status, error

A run summary (counts, timing) is also saved as RUN_SUMMARY in the key-value store. Fields the platform does not show, such as likes on some sites, stay null.

🚀 How to use (Apify Console)

1. Log in at console.apify.com and open Actors.
2. Find Video Transcript Scraper and open it.
3. Paste your video, playlist or channel links into the first field.
4. Set the maximum number of videos and, if you want, a language.
5. Click Start.
6. Watch the live log: each finished video is printed with its title, length, word count and language.
7. Open the Output tab, switch between the tables, and export to JSON, CSV or Excel.

🤖 Use via API and MCP

Start the Actor from the Apify API, a schedule or an integration such as Make, n8n or Zapier, then read the dataset items. For AI agents, add the Apify MCP server with claude mcp add --transport http apify "https://mcp.apify.com"; the agent finds this Actor through the server's search, details and call tools.

Best use cases

• Build a searchable archive of a channel or playlist.
• Feed transcripts into summarisation, Q&A or RAG pipelines.
• Quote exact moments using the segment jump links.
• Compare what different creators say about a topic across platforms.
• Collect captions in several languages for translation or research.

Frequently asked questions

Is it legal to get transcripts? The Actor reads public data that any visitor can see. You are responsible for following the rules of each site and the laws that apply to how you use the text.

Why did a video come back without a transcript? The video has no captions, is private, removed, age-restricted, region-locked or live, or the platform refused the request. The reason is in the error field and in the failed inputs table.

How do I get another language? Set language to a code such as es. If the video offers no captions in that language, the closest available one is returned and language_note explains it.

Can it transcribe audio? No. It uses captions that already exist on the video.

What happens when a site blocks the run? The run moves from no proxy to a datacenter proxy and then a residential proxy, retries on fresh residential IPs, and keeps the residential proxy for the rest of the run.

Why are likes or channel followers empty? Some platforms do not show them, so the field stays null.

Support and feedback

Found a link that should work but does not? Open an issue on the Actor page with the link and the run URL.