For developers
The YouTube transcript API that also does TikTok, Reels and X
One GET request with a video link. JSON back with the full text and timed segments, or a ready SRT, VTT or TXT file. No proxies, no headless browsers, no OAuth.
Quick start
Create a key in Dashboard → Developers, then:
curl "https://cliptranscript.app/v1/transcript?url=https://www.youtube.com/watch?v=jNQXAC9IVRw" \
-H "Authorization: Bearer cs_live_YOUR_KEY"Response
{
"platform": "youtube",
"id": "jNQXAC9IVRw",
"key": "youtube:jNQXAC9IVRw",
"url": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
"title": "Me at the zoo",
"author": null,
"language": "en",
"timed": true,
"duration": 18.88,
"words": 39,
"cached": true,
"text": "All right, so here we are, in front of the elephants ...",
"segments": [
{ "start": 1.2, "dur": 2.16, "text": "All right, so here we are, in front of the elephants" },
{ "start": 5.318, "dur": 2.656, "text": "the cool thing about these guys is that they have really..." }
],
"usage": { "used": 42, "limit": 1500, "period": "2026-11" }
}Parameters
| Name | Description |
|---|---|
url | Required. A YouTube (watch, youtu.be, Shorts), TikTok, Instagram (reel, p), Facebook (reel, videos, watch, fb.watch) or X (status) link. |
format | json (default), txt, srt or vtt. |
timestamps | With format=txt, 1 prefixes each line with [m:ss]. |
timed is true when segment times come from the platform (YouTube, TikTok and most Instagram Reels). When a platform only returns plain text (often Facebook and X), timed is false and segment times are estimated from speaking pace.
Errors
| Status | code | Meaning |
|---|---|---|
| 400 | invalid_url | Not a supported video link. |
| 401 | missing_key / invalid_key | Send Authorization: Bearer cs_live_…. |
| 402 | limit_reached / upgrade_required | Monthly allowance used, or no API plan. |
| 404 | no_transcript | The video has no speech or captions. Not counted. |
| 422 | unavailable | Private, removed or restricted video. Not counted. |
| 502 / 503 | upstream_busy / capacity | Retry after a short wait. Not counted. |
Python
import requests
r = requests.get(
"https://cliptranscript.app/v1/transcript",
params={"url": "https://www.tiktok.com/@washingtonpost/video/7609177768793787679"},
headers={"Authorization": "Bearer cs_live_YOUR_KEY"},
timeout=90,
)
r.raise_for_status()
for seg in r.json()["segments"]:
print(f"{seg['start']:7.2f} {seg['text']}")JavaScript
const res = await fetch(
"https://cliptranscript.app/v1/transcript?" + new URLSearchParams({ url: "https://www.instagram.com/reel/DHsD6HGqJhp/", format: "srt" }),
{ headers: { Authorization: "Bearer cs_live_YOUR_KEY" } },
);
const srt = await res.text();Usage
GET /v1/usage returns { plan, used, limit, period } for the key's account.
Comparing approaches (official Data API, open-source libraries, hosted APIs)? Read YouTube transcript API options for developers.