Universal Media Transcriber · Whisper Speech-to-Text

Turn any audio or video URL into text with OpenAI's Whisper (via the fast faster-whisper engine). Works with YouTube, podcasts, direct media files (mp3/mp4/wav/m4a…), and hundreds of other sites. Get plain text, timestamped segments, and ready-to-use SRT / VTT subtitles — in 90+ languages, with optional English translation. No API key. Pay per minute of audio.

Give it a URL, get a transcript. It handles the download, audio extraction, and transcription end to end.

Why this actor

• Any source — YouTube, podcast episodes, direct audio/video links, and hundreds of yt-dlp sites.
• 4 output formats — plain text, timestamped segments, SRT and WebVTT subtitles.
• 90+ languages — auto-detected, or set it explicitly. Optional translate-to-English.
• Word-level timestamps — perfect for captions and karaoke-style highlighting.
• Choose your model — tiny/base for fast & cheap, small/medium for higher accuracy.
• Bulk — pass many URLs in one run.
• Fair pricing — charged per minute of audio actually transcribed; failed items aren't charged.
• Schedule it — auto-transcribe new podcast episodes or channel uploads.

Use cases

You are a…	You use it to…
Podcaster / creator	Generate show notes, blog posts, and subtitles from episodes
Video team	Produce SRT/VTT captions for accessibility and SEO
Researcher / journalist	Transcribe interviews, lectures, briefings, calls
AI / RAG builder	Convert audio/video into text to index and search
Localization team	Translate foreign-language audio to English text

Input

Field	Type	Description
mediaUrls	array	Required. URLs of audio/video to transcribe.
model	string	tiny, base (default), small, or medium.
language	string	Two-letter code (e.g. en, es); blank = auto-detect.
translate	boolean	Translate speech to English instead of transcribing.
outputFormats	array	text, segments, srt, vtt.
wordTimestamps	boolean	Add per-word timestamps.
maxDurationMinutes	integer	Skip media longer than this (cost guard).
beamSize	integer	1 = fastest; higher = slightly more accurate.
proxyConfiguration	object	Proxy for downloads (helps with YouTube/geo limits).

Example — transcribe a YouTube video

{ "mediaUrls": ["https://www.youtube.com/watch?v=dQw4w9WgXcQ"], "model": "base", "outputFormats": ["text", "segments", "srt"] }

Example — transcribe podcast episodes with subtitles

{
  "mediaUrls": ["https://example.com/episode-42.mp3", "https://example.com/episode-43.mp3"],
  "model": "small",
  "outputFormats": ["text", "srt", "vtt"]
}

Output

One dataset item per URL:

{
  "url": "https://www.youtube.com/watch?v=...",
  "title": "Interview with ...",
  "uploader": "Some Channel",
  "source": "Youtube",
  "language": "en",
  "languageProbability": 0.99,
  "durationSeconds": 612.3,
  "durationMinutes": 10.21,
  "model": "base",
  "wordCount": 1580,
  "segmentCount": 142,
  "text": "Full transcript text…",
  "segments": [{ "start": 0.0, "end": 4.2, "text": "Hello and welcome…" }, ...],
  "srt": "1\n00:00:00,000 --> 00:00:04,200\nHello and welcome…\n\n..."
}

A run summary (URLs requested, transcribed, errors, audio-minutes charged) is saved to the key-value store under SUMMARY.

Pricing

Pay-per-event: charged per minute of audio successfully transcribed (rounded up), plus a small per-run start fee. Failed downloads/transcriptions are not charged. Larger models (small, medium) cost more compute, so choose the model that fits your accuracy/budget. See the Pricing tab for the live per-minute rate.

Tip: base is the sweet spot for most content. Use tiny for quick drafts, medium for hard audio.

FAQ

What sources work? YouTube, podcast episodes, direct audio/video file URLs, and hundreds of sites supported by yt-dlp. For best results with YouTube at scale, enable Apify Proxy.

Which languages? Whisper supports 90+ languages and auto-detects by default. translate: true outputs English.

How accurate is it? Whisper is state-of-the-art. Accuracy scales with model size; base is great for clear speech, small/medium handle accents, noise and jargon better.

Can I get subtitles? Yes — include srt and/or vtt in outputFormats.

Can I auto-transcribe new episodes? Yes — schedule the actor over a list/feed of new URLs.

Limitations

• Transcription is compute-bound; long media takes longer (and costs more minutes). Use maxDurationMinutes to cap.
• Some sites (certain social platforms) may rate-limit or block downloads; use a proxy and expect best-effort on those.
• Very noisy or overlapping speech is inherently harder; try a larger model.