YouTube Transcript Pro — Office Edition

Bulk YouTube transcript & subtitle extractor for content teams, research, and AI training datasets.

Features

• Bulk URL processing (watch, youtu.be, shorts, embed)
• 100+ languages
• Optional timestamped output ([mm:ss] text)
• Job tracking — every run gets a unique job_id and optional job_label
• Per-video stats: line count, word count, duration
• Run summary saved to key-value store as JOB_SUMMARY

Run limits (Office edition)

To keep runs fast and lightweight, this edition caps each execution:

• Max 5 videos per run — extra URLs are trimmed with a log notice.
• Max 3 minutes runtime — any video beyond this is skipped and reported in the summary as stopped_by_timeout: true.

For larger bulk jobs, split the input across multiple runs.

Input

Field	Type	Description
urls	array (required)	YouTube video URLs
language	string	Preferred subtitle language code (default: en)
jobLabel	string	Optional label for this batch (team/project name)
includeTimestamps	boolean	Prefix each line with [mm:ss]

Output (per video)

{
  "job_id": "JOB-20260405XXXXXX-AB12",
  "job_label": "marketing-q2",
  "video_id": "dQw4w9WgXcQ",
  "url": "https://...",
  "language": "en",
  "line_count": 142,
  "word_count": 980,
  "duration_sec": 213,
  "extracted_at": "2026-04-05T...Z",
  "transcript": "...",
  "lines": ["..."],
  "raw": [...]
}

Run Summary

After every run, a JOB_SUMMARY record is written to the default key-value store containing totals, success/failure counts, and timing.