Twitter Transcript API

The Twitter Transcript API extracts a full spoken transcript from a video embedded in a tweet, including timestamped segments, the full transcript text, and word count.

Endpoint

https://api.socialkit.dev/twitter/transcript

Example Request

GET https://api.socialkit.dev/twitter/transcript?access_key=<your-access-key>&url=https://x.com/OpenAI/status/1803543210987654321

Response

{
	"success": true,
	"data": {
		"url": "https://x.com/OpenAI/status/1803543210987654321",
		"transcript": [
			{
				"text": "Today we're announcing GPT-5, our most capable model yet.",
				"start": 0,
				"duration": 4.2
			},
			{
				"text": "It features significantly improved reasoning and can handle complex multi-step tasks.",
				"start": 4.2,
				"duration": 5.1
			},
			{
				"text": "GPT-5 is available right now in ChatGPT and through our API.",
				"start": 9.3,
				"duration": 4.8
			}
		],
		"text": "Today we're announcing GPT-5, our most capable model yet. It features significantly improved reasoning and can handle complex multi-step tasks. GPT-5 is available right now in ChatGPT and through our API.",
		"wordCount": 37
	}
}

Parameters

url string Required
The URL of a tweet containing a video (e.g. https://x.com/OpenAI/status/1803543210987654321).

access_key string Required
Your API access key. Can be provided via the access_key query parameter, x-access-key header, or request body.

cache boolean Optional Defaults to false
Cache the response for faster subsequent requests.

cache_ttl number Optional Defaults to 2592000
Cache the response for a custom TTL (in seconds). Maximum 2592000 seconds (1 month), minimum 3600 seconds (1 hour).

Response Fields

The API returns the following data fields:

• url: The original tweet URL provided
• transcript: Array of timestamped transcript segments
  • text: The spoken text for this segment
  • start: Start time of the segment in seconds (number)
  • duration: Duration of the segment in seconds (number)
• text: The full transcript as a single concatenated string
• wordCount: Total number of words in the transcript (number)

Use Cases

• Content Repurposing: Turn video tweets into blog posts, threads, or summaries
• Accessibility: Generate captions or subtitles for Twitter video content
• Media Monitoring: Transcribe and index video statements from public figures
• Research: Extract spoken claims from interviews or announcements shared on Twitter/X
• SEO: Index spoken content from social video to improve content discoverability
• Fact Checking: Pull verbatim quotes from video tweets for verification workflows
• AI Pipelines: Feed transcripts into summarization or classification models

Notes

• Only tweets containing a video attachment are supported; image-only or text-only tweets will return an error
• Transcript accuracy depends on audio clarity and language — English is best supported
• The start and duration fields are in seconds and can be used to build subtitle files (e.g. SRT, VTT)
• Both https://x.com/ and https://twitter.com/ tweet URL formats are supported