X video transcript API
AI transcription of the video in a post.
100 free credits. No card needed.
x.com/coinbase/st...
Sent with the link
- Cost
- 1 credit
- Fields
- 11
- Cached
- 30 days
One transcript a request
What you get in each transcript
What you can build with it
A video post's speech as text. Pass a post with a video and get its speech as text with timed segments, for search, quotes or a summary. The sample's 45 second video comes back as one segment.
- source
- ai
- duration_seconds
- 45
- language
- en
Read from the audio every time. The transcript is made from the audio of the post's first video, so source always reads ai, and the language is the one heard in it.
- video_id
- 2111731581702598777
- source
- ai
A budget for long clips. A video up to 2 minutes costs 1 credit. Past that it needs allow_long_transcription=true and costs 5 credits per started minute of the whole video, so a 3 minute clip costs 15.
- duration_seconds
- 45
- language
- en
From a key to JSON.
url (x.com/coinbase/status/206697087...).data.The request. GET /v1/twitter/tweet/transcript
curl -G "https://api.socialscrape.dev/v1/twitter/tweet/transcript" \ --data-urlencode "url=https://x.com/coinbase/status/2066970874016645585" \ -H "x-api-key: $SOCIALSCRAPE_API_KEY"
import requests r = requests.get( "https://api.socialscrape.dev/v1/twitter/tweet/transcript", params={ "url": "https://x.com/coinbase/status/2066970874016645585", }, headers={"x-api-key": KEY}, ) transcript = r.json()["data"]
const url = new URL("https://api.socialscrape.dev/v1/twitter/tweet/transcript"); url.searchParams.set("url", "https://x.com/coinbase/status/2066970874016645585"); const res = await fetch(url, { headers: { "x-api-key": KEY }, }); const { data } = await res.json();
The response. A sample, in the shape every call returns.
{
"ok": true,
"credits_charged": 1,
"data": {
"text": "Example transcript text. This is...",
"duration_seconds": 45,
"url": "https://x.com/yashsrivasta7a/stat...",
"video_id": "2111731581702598777",
"language": "en",
"source": "ai",
"segments": [ ... ]
}
}
Ask in plain words. Ask your agent for X video transcript in plain words, and the SocialScrape MCP server makes the call.
Get the X video transcript for x.com/coinbase/status/206697087...
Here is the transcript: "Example transcript text. This is...".
One tool call
X video transcript, given the link.
Same price as the API
1 credit a call; a repeat within 30 days is free.
Runs on your machine
Your key stays in your agent's settings.
Pay only for data you get.
Failed requests and cached repeats are free.
1,000 credits
$13 one time
Get an API key- $13.00 per 1,000 credits
- Credits never expire
- No subscription
25,000 credits
$55 one time
Get an API key- $2.20 per 1,000 credits
- Credits never expire
- No subscription
230,000 credits
$349 one time
Get an API key- $1.52 per 1,000 credits
- Credits never expire
- No subscription
Questions, answered.
Common questions
Which X posts can be transcribed?
Posts with a video. A post without one, or with an animated GIF, which has no audio, answers transcript_unavailable, free.
What does a long X video answer without allow_long_transcription?
invalid_params, with the cost named in the message, and nothing is charged.
What does it return?
Videos over 2 minutes need allow_long_transcription=true and cost 5 credits per started minute of the whole video. Posts without video, and animated GIFs, answer transcript_unavailable (free). The transcript has up to 11 fields, and the sample fills video_id, url, language, source, duration_seconds, text and segments.
What do I send?
Send one of url or id.
url: Post URL (x.com/<user>/status/<id>).
id: Post id.
allow_long_transcription: Transcribe audio past 2 minutes with AI, at 5 credits per started minute of the whole video. Without it a longer video answers invalid_params naming the cost, which is free.
Which errors can it return?
invalid_params, unsupported_url, not_found and transcript_unavailable. None is charged, and the errors list says which to retry.
Is there an SDK call for it?
client.twitter.tweet_transcript(...) in Python and client.twitter.tweetTranscript() in TypeScript.
Start pulling social data.
100 free credits. No card needed.