Transcripts¶

The YouTube Toolkit provides two transcript extraction flows: a synchronous single-video endpoint and an async batch endpoint for processing multiple URLs in one job.

Transcript endpoints¶

REST API Examples¶

Get a transcript¶

curl "https://youtube.toolkitapi.io/v1/youtube/transcript?url=https://www.youtube.com/watch?v=dQw4w9WgXcQ" \
  -H "X-API-Key: YOUR_KEY"

const params = new URLSearchParams({ url: "https://www.youtube.com/watch?v=dQw4w9WgXcQ" });
const resp = await fetch(`https://youtube.toolkitapi.io/v1/youtube/transcript?${params}`, {
  headers: { "X-API-Key": "YOUR_KEY" },
});
const data = await resp.json();
console.log(data.transcript.substring(0, 200));

Batch transcript request¶

curl -X POST "https://youtube.toolkitapi.io/v1/youtube/transcript/batch" \
  -H "X-API-Key: YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://www.youtube.com/watch?v=dQw4w9WgXcQ", "https://www.youtube.com/watch?v=abc123"]}'

Python SDK examples¶

Single transcript¶

from toolkitapi import Media

with Media(api_key="tk_...") as media:
    result = media.youtube_transcript("https://www.youtube.com/watch?v=dQw4w9WgXcQ")
    print(result["lang"], len(result["content"]), "segments")

Request a specific language¶

from toolkitapi import Media

with Media(api_key="tk_...") as media:
    result = media.youtube_transcript(
        "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
        lang="es",
    )
    for segment in result["content"]:
        print(f"[{segment['offset']}ms] {segment['text']}")

Submit a batch job¶

from toolkitapi import Media

urls = [
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
    "https://www.youtube.com/watch?v=9bZkp7q19f0",
    "https://youtu.be/kffacxfA7G4",
]

with Media(api_key="tk_...") as media:
    job = media.youtube_transcript_batch({"urls": urls, "lang": "en"})
    print(job["jobId"])

Poll batch job status¶

import time
from toolkitapi import Media

with Media(api_key="tk_...") as media:
    job_id = "abc-123"
    while True:
        status = media.youtube_transcript_batch_status(job_id)
        if status["status"] in ("completed", "failed"):
            break
        time.sleep(2)

    for item in status["results"]:
        if item["status"] == "succeeded":
            print(f"{item['id']}: {len(item['content'])} segments")
        else:
            print(f"{item['id']}: FAILED — {item['error']}")

Transcript response fields¶

content[] segment fields¶

Batch job fields¶

Submit response (POST /transcript/batch)¶

Status response (GET /transcript/batch/{job_id})¶

results[] item fields¶

Tips¶

• The lang parameter is a preference — if the requested language is unavailable the API returns the closest available language. Check the lang field in the response to confirm what was actually returned.
• availableLangs lists all auto-generated and manually added captions for the video.
• Batch jobs expire after a short window — poll promptly after submission and persist results on your end.
• For AI/RAG pipelines, the content array gives you clean, timestamped segments ready for chunking and embedding.
• The batch endpoint accepts up to 20 URLs per job. For larger sets, split into multiple jobs and submit concurrently.

Jump straight to live tools¶

• https://youtube.toolkitapi.io/tools/youtube/youtube-transcript/
• https://youtube.toolkitapi.io/tools/youtube/youtube-transcript-batch/