Twitter Transcript Scraper

What is Twitter Transcript Scraper?

Twitter Transcript Scraper turns X (Twitter) videos into text. Paste post links or usernames, click Start, and get one row per video with everything that is said in it: the full transcript, the transcript in timed parts and the spoken language, next to the post's text, author, length, views, likes, reposts and replies. Download it as Excel, CSV or JSON, or schedule it to transcribe an account's new videos every day.

• 🗣️ Full transcripts: every word spoken in the video, as one text and in timed parts
• 🌍 Any language: the spoken language is detected for each video, and the transcript is in that language
• 👤 Whole accounts or single posts: the newest videos of any public account, or any video by its post link
• 🔁 Reposts and quotes handled: a repost gives the original video, a post that quotes a video gives the quoted video
• 🎵 No-speech videos are free: music-only clips, GIFs and posts without a video are saved and marked, and you are not charged for them
• 🔓 No login: no X account, cookies or API key needed

What can Twitter Transcript Scraper do?

• Transcribe the newest videos of many accounts in one run, up to the number you choose per account
• Transcribe single videos from their post links (x.com or twitter.com), or from bare post IDs
• Find out what people say on camera, not just what they write in the post
• Get timed transcripts to find the moment something is said, quote it or cut a clip
• Export to Excel, CSV, JSON, HTML or XML, or send the results to Google Sheets, Make, Zapier and more

What data can you get from X videos?

🗣️ Transcript	⏱️ Timed transcript parts	🌍 Spoken language
🔗 Post link and ID	✍️ Post text and language	📅 Posting date
⏱️ Video length in seconds	👁️ Views, ❤️ likes, 🔁 reposts, 💬 replies, 🔖 bookmarks	👤 Author username, name, followers, checkmark
📹 Video file link	🖼️ Thumbnail	🎞️ Video or GIF

How to get the transcript of a Twitter video

1. Create a free Apify account (no credit card needed).
2. Open Twitter Transcript Scraper.
3. Paste one or more post links, or add usernames to transcribe their newest videos.
4. Choose how many videos per profile you want transcribed.
5. Click Start, then download the results as Excel, CSV, JSON, HTML or XML.

How much does it cost to transcribe Twitter videos?

You pay per video with speech, whether its transcript comes from the video's own captions or from its sound. The transcript price includes the first minute of each video; a longer video also pays an extra minute for every started minute past the first. Higher Apify plans pay less:

Apify plan	Per 1,000 transcripts	Per 1,000 extra minutes
Free	$3.50	$2.50
Starter (Bronze)	$3.00	$2.20
Scale (Silver)	$2.80	$2.00
Business (Gold) and higher	$2.50	$1.80

So a 5-minute video is 1 transcript and 4 extra minutes. Platform usage is included in these prices, so you pay nothing else for the run. There is no start fee.

Example: 1,000 one-minute videos cost $3.00 on Starter, $2.80 on Scale and $2.50 on Business and higher.

Never charged:

• videos with no speech (music only, sounds or silence), GIFs and posts without a video,
• videos longer than your Longest video (minutes),
• links that do not exist and videos whose sound could not be downloaded or transcribed,
• extra minutes of any video that a Free plan run cut to its first minute, and videos the run reached after its free minute was used.

Every Apify account gets $5 of free usage each month.

On the Apify Free plan, transcripts made from the audio use a lighter, slower method, so expect lower accuracy (more wrong words, most of all in songs and noisy videos), and a run transcribes up to 1 minute of audio: a longer video gets its first minute, and videos past that minute are saved as not transcribed and are not charged. Keep the default 1 GB memory: with less, a Free plan run transcribes no audio. Transcripts from a video's own captions are the same on every plan. Paid Apify plans get full-accuracy transcripts with no such limit.

⬇️ Input

Setting	What it does
Post links	Posts with a video, one per line: https://x.com/NASA/status/2115845108176441635, a twitter.com link, a link ending in /video/1, or a bare post ID
Profiles	Usernames or profile links (NASA, @NASA, https://x.com/NASA): each gives its newest videos
Videos per profile	The most videos to transcribe from each profile, newest first (default 10)
Longest video (minutes)	Longer videos are saved with their post details but not transcribed, and cost nothing (default 60)
Spoken language (optional)	A code such as en, es or fr when every video is in that language. Leave empty to detect each video's language

Example:

{
  "postUrls": ["https://x.com/NASA/status/2115845108176441635"],
  "profiles": ["SpaceX"],
  "videosPerProfile": 10
}

A single link in videoUrl works too, as in other X transcript tools.

⬆️ Output

You get one row per video. View them as a table in Apify Console or download them as Excel, CSV, JSON, HTML or XML.

{
  "id": "2115845108176441635",
  "url": "https://x.com/NASA/status/2115845108176441635",
  "authorUsername": "NASA",
  "authorName": "NASA",
  "text": "@csa_asc @Space_Station Solar array, roll out! At about four-and-a-half hours into today’s spacewalk, the solar array began deployment. It is now fully deployed.",
  "createdAt": "2026-10-29T16:35:58Z",
  "durationSeconds": 22.1,
  "transcriptStatus": "transcribed",
  "transcriptSource": "speech",
  "transcriptLanguage": "en",
  "transcript": "Four hours and 23 minutes into today's spacewalk, the seventh episode. ISS rollout solar array begins to roll out. and will for the next 10 minutes or so to a fully deployed length of 60 feet",
  "transcriptSegments": [
    {"start": 3.54, "end": 9.26, "text": "Four hours and 23 minutes into today's spacewalk, the seventh episode."},
    {"start": 10.32, "end": 14.8, "text": "ISS rollout solar array begins to roll out."},
    {"start": 14.91, "end": 22.11, "text": "and will for the next 10 minutes or so to a fully deployed length of 60 feet"}
  ],
  "videoId": "2115844986121973760",
  "videoType": "video",
  "videoUrl": "https://video.twimg.com/amplify_video/2115844986121973760/vid/avc1/3840x2160/bFtWNhBlnRTuKsG-.mp4?tag=29",
  "thumbnailUrl": "https://pbs.twimg.com/amplify_video_thumb/2115844986121973760/img/UVq5lw5EUQLmo5YD.jpg",
  "lang": "en",
  "authorId": "11348282",
  "authorVerified": true,
  "authorFollowers": 92296242,
  "views": 159119,
  "likes": 316,
  "retweets": 25,
  "replies": 14,
  "quotes": 5,
  "bookmarks": 9,
  "fromQuotedPost": false,
  "source": "https://x.com/NASA/status/2115845108176441635",
  "scrapedAt": "2026-11-01T13:09:47Z"
}

transcriptStatus tells you what happened with each video:

Status	Meaning	Charged
transcribed	Speech was found; transcript has it	Yes
no_speech	Music, sounds or silence only; transcript is empty	No
gif	A GIF, which has no sound	No
too_long	Longer than Longest video (minutes)	No
no_video	The post link has no video	No
audio_unavailable	The video's sound could not be read this time; run it again later	No

transcriptSource says where the words came from: captions when the video has its own captions in its spoken language, speech when they were taken from what is said. Each run also saves a short report: how many videos had speech, how many videos each profile has, and any link that does not exist.

How can I use Twitter video transcripts?

• Media monitoring: get the exact words of news clips, press conferences, speeches and interviews posted on X
• Brand and competitor research: track what founders, brands and politicians say on camera
• Content repurposing: turn video posts into threads, articles, newsletters and show notes
• Research and fact-checking: quote what was actually said, with the second it was said
• Search and analysis: load transcripts into spreadsheets, databases or search tools, or summarize them in your own workflow

Use it from your code

Run the actor from your own code with the Apify API or the Python and JavaScript clients:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("automly/x-twitter-video-transcript-scraper").call(run_input={
    "profiles": ["NASA"],
    "videosPerProfile": 10,
})
for video in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(video["url"], video["transcriptLanguage"], video["transcript"])

Use with assistants and agents (MCP)

Connect any MCP-compatible assistant or agent to the Apify MCP server at https://mcp.apify.com (sign in with your Apify account), and it can find this actor, fill in the input and read the results by itself. Ask in plain words, for example "Get the transcripts of NASA's last 10 videos on X", and the agent runs automly/x-twitter-video-transcript-scraper with an input like {"profiles": ["NASA"], "videosPerProfile": 10}. Agent runs cost the same as runs you start yourself.

❓FAQ

How accurate are the transcripts?

Clear speech comes out close to word for word. Transcripts made from the spoken words have punctuation; a video's own captions keep the punctuation they have, which is sometimes none. Accuracy drops with loud background music, several people talking at once and heavy accents. Words sung in a song are transcribed like speech.

Which languages are supported?

The spoken language is detected for each video, and the transcript is in that language (it is not translated). English, Spanish, French, German, Portuguese, Arabic, Hindi, Japanese, Korean, Chinese and many other languages work. If all your videos are in one language, set Spoken language to its code (for example es).

Is the transcript the same as the post text?

No. The post text is what the author wrote. The transcript is what is said in the video. You get both.

How many videos can I get from one account?

X shows visitors about the latest 650-850 posts of an account, so you get the videos among them, newest first, up to Videos per profile. For some big accounts X shows visitors only their few newest posts and their most-liked posts (about 100); you then get the videos among the newest ones first, then among the most-liked ones, and the run report says so. Reposts of other accounts' videos are skipped; add their post links to transcribe them. The run report says how many videos each profile had when there were fewer than you asked for.

Does it work with long videos?

Yes. Interviews, podcasts and streams of an hour or more work. Longest video (minutes) is 60 by default, so you never pay for a three-hour video by accident; raise it when you want long videos transcribed.

What happens with music-only videos and GIFs?

Videos with only music, sound effects or silence are saved with transcriptStatus set to no_speech and an empty transcript. GIFs have no sound and are saved as gif. You are not charged for either.

What happens with a post that quotes a video?

If the post has no video of its own and quotes a post with a video, the quoted video is transcribed and fromQuotedPost is true. A repost gives the original post's video. The same video is transcribed and charged once per run, even when it is in several posts.

What happens with a link that does not exist?

It is skipped and listed in the run's report as not found. The run still finishes successfully and you are not charged for it. Spaces and live broadcasts are not supported.

Do I need an X account?

No. You don't need an X account, cookies or an API key. Only public posts are read: posts of protected accounts are not available.

Can I connect Twitter Transcript Scraper to other tools?

Yes. Start runs and download results with the Apify API or the Python and JavaScript clients, send results to Make, Zapier, n8n, Google Sheets or Slack with Apify integrations and webhooks, or let agents run it through the Apify MCP server at mcp.apify.com.

Is it legal to transcribe Twitter videos?

This tool only processes videos that people chose to show publicly. It does not collect private data such as email addresses or phone numbers. Videos and their transcripts can still contain personal data, which may be protected by laws such as GDPR, and videos are protected by copyright, so use the transcripts for a legitimate reason such as research, and ask a lawyer if you are unsure. You can read more in Is web scraping legal?

Twitter Transcript Scraper is an independent tool. It is not affiliated with, endorsed by or sponsored by X Corp.

Something isn't working. What should I do?

Open the Issues tab and tell us the link and what you expected. We usually reply within a day. If the tool helped you, a review on its Store page helps other people find it.

Scrape more X (Twitter) data

Actor	What you get
Twitter Scraper	An account's posts with text, views, likes, reposts, replies and media, plus its profile
Tweet Scraper	Any posts by link or ID, with their counts, media, links and quoted post
Twitter Profile Scraper	Profiles: followers, following, post count, bio, location, website, join date