The transcript API for social media video.

A Python request is sent to the transcript API. The API resolves the video and returns timestamped transcript segments as JSON.

import requests

response = requests.post(
  "https://transcriptfetch.com/api/v2/transcripts/video",
  headers={"Authorization": "Bearer tf_live_••••"},
  json={
    "video": "https://youtu.be/bNpx7gpSqbY",
    "timestamps": True
  }
)

transcript = response.json()["data"]

{
  "ok": true,
  "request_id": "req_7H2b9YQk",
  "data": {
    "kind": "transcript",
    "platform": "youtube",
    "video_id": "bNpx7gpSqbY",
    "duration": 401,
    "language": "en",
    "segments": [
      { "start": 0.0, "text": "Today we're looking at..." },
      { "start": 4.2, "text": "The first thing to know..." },
      { "start": 9.8, "text": "Each layer learns..." },
      { "start": 15.6, "text": "That signal carries..." },
      ... 45 more segments
    ]
  },
  "usage": { "credits_spent": 1, "balance": 4812 }
}