Smart Transcript — YouTube Transcriber

Transcribe the speech from up to 3 YouTube videos in one run, even when the videos have no captions. Choose the exact video links you want, or select videos from a playlist or channel.

What you get: full transcription text, detected language, timestamped segments when available, and JSON results for each video. Choose the spoken language or use automatic detection.

How to use

1. Add 1–3 YouTube video links to Video URLs (up to 3). Videos can come from different channels.
2. Select Transcribe and choose a spoken language, or leave it on Auto detect.
3. Click Start. Open Results / Dataset to view or export the output.

To discover videos from a playlist or channel, use YouTube URL instead and choose Maximum videos from playlist/channel (1–3). Do not fill in both URL fields.

Example: three videos you choose

{
  "sourceUrls": [
    "https://www.youtube.com/watch?v=jNQXAC9IVRw",
    "https://www.youtube.com/watch?v=9UkCsZUrZ2g",
    "https://www.youtube.com/watch?v=3A5vQ-LBKfU"
  ],
  "language": "auto",
  "mode": "submit"
}

Each link selects one exact video. For playlists and channels, up to three videos are selected in discovery order, which may change over time.

Languages

Choose Auto detect to identify the spoken language, or select it manually in the Language field. The form offers 37 language selections, including separate options for Portuguese (Brazil) and Portuguese (Portugal).

Supported languages include Arabic, Bengali, Chinese, Czech, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Malay, Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Spanish, Swahili, Swedish, Tagalog (Filipino), Tamil, Thai, Turkish, Ukrainian, Urdu, and Vietnamese.

Transcription stays in the original spoken language. Translation and speaker identification are not included. Accuracy depends on the audio quality and language.

Results and rejected videos

Each requested video receives a separate Dataset row:

• completed — transcription text and metadata are available.
• rejected — that video could not be accepted, for example because it exceeds 60 minutes or its duration cannot be verified.

One rejected video does not cancel valid videos. A run can finish successfully with both completed and rejected items. Check each row's status to see what happened. Rejected videos are not charged for transcription.

Price and limits

• US$0.03 per started minute of successfully delivered transcription, rounded up separately for each completed video.
• Up to 3 videos per run and 60 minutes per video.
• Works with speech audio, even without YouTube captions. Audio quality affects accuracy.
• Live streams, Shorts, and unsupported or inaccessible videos are not covered by this service.
• Processing time depends on video length and service availability. Platform usage charges may apply separately.

Price examples: 30 seconds = US$0.03; 7 minutes 19 seconds = US$0.24; 12 minutes 52 seconds = US$0.39. Rejected videos = US$0 in transcription charges.

Use the API

Start a run with your Apify API token (keep it private):

curl -X POST 'https://api.apify.com/v2/actors/QPgSQoYb4m8AjQco0/runs' \
  -H 'Authorization: Bearer YOUR_APIFY_TOKEN' \
  -H 'Content-Type: application/json' \
  -d '{"sourceUrls":["https://www.youtube.com/watch?v=jNQXAC9IVRw"],"mode":"submit","language":"auto"}'

The response includes a Run ID and Dataset ID. After the run completes, fetch the results:

GET https://api.apify.com/v2/datasets/YOUR_DATASET_ID/items?format=json

For questions or errors, use the Issues tab and include your Run ID. Submit only content you have permission to process.