$5 free credits when you sign up Claim now
Marlin 2b now available Test it!
MiniMax H3 now available! Test it!
Sign in Get started Claim $5
Browse all Models
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all
Browse all

YouTube Transcript API With or Without Captions

Send a YouTube link and get the transcript as JSON with timestamps. deAPI downloads the video and runs Whisper CT2 on the audio, so you don't run proxies and the video doesn't need captions. Speaker labels and video metadata are one flag each.

Get $5 Free Credits Read the Docs

$5 in free credits to start. No credit card required.

per hour of video, segment timestamps
$0.052
per hour with speaker labels or word timestamps
$0.078
free credits, about 96 hours of video
$5

Why youtube-transcript-api Gets IP Blocked on Cloud Servers

Most YouTube transcript tools read the captions YouTube already has. That is cheap, and it fails in production in two ways.

  • Blocked from cloud servers

    The youtube-transcript-api README says YouTube blocks most IPs that belong to cloud providers such as AWS, Google Cloud and Azure, which leads to RequestBlocked or IpBlocked errors. With deAPI the request to YouTube comes from our side.

  • No captions, no transcript

    Caption APIs return nothing, or charge more for AI transcription, when a video has no captions. deAPI always transcribes the audio, so a YouTube video costs the same per hour, captions or not.

Request and JSON Response

One POST to /api/v2/audio/transcriptions. The call returns a request ID, and the result arrives on the job endpoint or your webhook.

Request
source_url
A public YouTube link. Age- or region-restricted videos fail. X, Twitch, Kick and TikTok links work too.
include_ts
Set to true to get timestamps. Without it, ts_level is ignored.
ts_level
Timestamps per segment (free) or per word (×1.5).
diarize
Label who is speaking (×1.5, not stacked with word timestamps).
include_metadata
Title, channel, description, duration, views and likes (+$0.005).
webhook_url
Called when the job starts, finishes or fails. Signed with HMAC-SHA256 when you set a webhook secret.
Response (JSON)
text
The full transcript.
language
Detected language of the audio.
segments
Start, end and text for each segment, plus the speaker when diarization is on.
metadata
Video details when requested.

YouTube Transcript API Pricing Compared

List prices from each vendor's pricing page and docs, checked 27 October 2026. Caption APIs charge per transcript, so their hourly cost depends on video length.

Service How it gets the text Price as listed Video without captions
deAPI Whisper CT2 on the audio $0.052 per hour Same price
ElevenLabs Scribe v2 Speech-to-text from a YouTube link $0.22 per hour Same price
Supadata (Pro, billed monthly) YouTube captions, AI fallback $17/mo for 3,000 credits, 1 credit per transcript 2 credits per minute, ≈ $0.68 per hour
youtube-transcript.io (Plus) YouTube captions $9.99/mo for 1,000 transcripts No AI transcription offered
deAPI on Apify · YouTube Transcript AI Speech recognition on the audio $0.0075 per video + $0.00106 per minute, $0.0711 for one hour Same price
Apify actor agentx/youtube-transcript Speech recognition on the audio $0.3483 per video on Apify Gold and up ($0.387 on Free), plus run costs Same price
youtube-transcript-api (open source) YouTube captions Free, plus proxy costs No transcript

deAPI price from the /price endpoint on 27 October 2026 for a one-hour video, segment timestamps. Check the current amount with /price before each job. The deAPI actor on Apify needs no deAPI key and bills through Apify; see deAPI on Apify.

Rate Limits, Spend Caps and Data Handling

  • Rate limits

    Free account: 1 request per minute and 10 per day. Limits go up after the first payment. For more, write to [email protected].

  • Know the cost first

    Every task has a /price endpoint. Quote a video before you queue it.

  • Cap each key

    Give a key a spend limit per day, week or month, an expiry date and an IP allowlist. A key over its limit gets a 402 and the job does not run.

  • Prepaid credit

    You top up credit and jobs draw from it. A key never spends more than the account balance.

  • Webhooks

    Get job.completed or job.failed on your endpoint instead of polling.

  • Data handling

    Generated files are deleted no later than 24 hours after job creation.

  • $5 in free credits
  • No credit card required
  • Prepaid, no subscription

Quote a Batch of Videos, Then Run It

$5 in credits covers about 96 hours of video, at up to 10 requests a day until your first payment. Email verification, no card. For higher volumes, write to [email protected].

Get $5 Free Credits Talk to Sales

See also Speech-to-Text API Pricing, Transcription MCP Server and deAPI on Apify.

Frequently Asked Questions

YouTube transcripts, limits and pricing

Yes. deAPI runs Whisper CT2 on the audio track, so you get a transcript for videos with auto-captions, manual captions or none at all.
Send the link to deAPI. We fetch the video on our side, so your servers never request YouTube directly and you don't need proxies.
$0.052 for a one-hour video with segment timestamps. Word-level timestamps or speaker labels multiply the price by 1.5, once, even with both on: $0.078 per hour. Video metadata adds $0.005 per request. The /price endpoint returns the exact amount before you run the job.
Free account: 1 request per minute and 10 per day. Limits go up after the first payment. For more, write to [email protected].
The API returns JSON, not SRT or VTT files. Each segment has a start and end time, which converts to either format in a few lines.
Yes, through the deAPI MCP server in Claude, Cursor, ChatGPT and other MCP clients. See Transcription MCP Server.