X (Twitter) Video Transcript Scraper - Video to Text

Turn any X (Twitter) video into text. Paste post links, up to 500 per run, and get the full transcript of each video with timestamps and the detected language, plus the post's author, text, date and engagement. Works for clips, interviews, podcasts and long videos up to 60 minutes. No login and no X API key needed.

What you get

One row per post with:

• Transcript: the full spoken text in transcript, plus wordCount and the detected language
• Timestamps: a segments list with the start and end time (in seconds) of every sentence, ready for subtitles, clipping or quotes
• Post details: post ID and URL, author username and display name, post text, date, views, likes, reposts and replies
• Video details: duration and thumbnail. Posts that quote a video are supported too (videoFromQuotedTweet: true)

Use cases

• Media monitoring: get the exact words from news clips, press conferences and political speeches posted on X
• Content repurposing: turn video posts into threads, articles, newsletters and show notes
• Research and fact-checking: quote what was actually said, with timestamps
• AI and LLM workflows: feed clean transcripts into ChatGPT, Claude or your own model for summaries, sentiment and insights
• Podcast and interview clips: search and archive what guests said
• Subtitles and accessibility: export timestamped segments as captions

How to use

1. Paste X or Twitter post links (or post IDs), one per line
2. Optionally set the Spoken language if you know it (it's detected automatically otherwise)
3. Run the actor and open the Dataset tab
4. Export to CSV, Excel, JSON or Google Sheets

Output example

{
  "inputUrl": "https://x.com/TheoVon/status/1916982720317821050",
  "status": "ok",
  "tweetId": "1916982720317821050",
  "tweetUrl": "https://x.com/TheoVon/status/1916982720317821050",
  "videoFromQuotedTweet": false,
  "author": "TheoVon",
  "durationSeconds": 84.3,
  "views": 166314,
  "likes": 909,
  "language": "en",
  "transcript": "Since you're kind of like a leader in innovation and technology in our world, you know? Um, do you, how do you know that what your convictions are? ...",
  "segments": [
    { "start": 0.001, "end": 10.711, "text": "Since you're kind of like a leader in innovation and technology in our world, you know?" },
    { "start": 10.711, "end": 20.911, "text": "Um, do you, how do you know that what your convictions are? How do you gauge if what your convictions are are the best for everybody kind of?" }
  ]
}

Pricing

• Transcript: a small fee per video transcribed
• Transcript minute: a fee per minute of video (rounded up to the next minute)

Posts without a video, GIFs, videos with no spoken words, deleted posts and invalid links are free. They appear as rows with a status and a message explaining why.

Free Apify plans can transcribe up to 5 videos per run (3 runs a day). Upgrade to a paid plan to remove these limits.

FAQ

Which languages are supported? Most spoken languages. The language is detected automatically and returned in language.

How long can videos be? Up to 60 minutes per video.

Can I transcribe a post that quotes a video? Yes. If the post itself has no video but quotes one, the quoted video is transcribed and videoFromQuotedTweet is true.

Why is a post marked no_speech? The video only contains music or sound effects, so there is nothing to transcribe. You are not charged.