Video & Audio Transcriber — Speech to Text, Subtitles & SRT

Turn any video or audio URL into accurate text — with language detection, timestamped segments, ready-to-use SRT/WebVTT subtitles, and an optional AI summary. Paste links from YouTube, TikTok, Facebook, X (Twitter), Vimeo, SoundCloud, podcast RSS feeds and 1000+ other sites, or point it straight at a media file. Export to CSV, JSON or Excel.

Powered by OpenAI Whisper speech-to-text. No login, no browser extension, no manual downloading.

What it does

• 🎙️ Speech to text — a clean, full transcript for every URL.
• 🌍 Any language — automatic language detection, or pass a language hint.
• ⏱️ Timestamps — optional segment-level start/end times.
• 💬 Subtitles — ready-to-use SRT and WebVTT files for captioning and editing.
• 🧠 AI summary — optional 1–2 sentence summary plus keywords per item.
• 🔗 Works everywhere — YouTube, TikTok, Facebook, Instagram, X, Vimeo, SoundCloud, Dailymotion, podcasts (RSS), and 1000+ sites, or direct media files (mp3, mp4, wav, m4a, flac, ogg, webm, mov).
• 📄 Export — CSV, JSON, Excel via the Apify dataset.

Input

Field	Type	Description
mediaUrls	array	Required. Video/audio page URLs or direct media-file links, one per line.
includeSummary	boolean	Add an AI summary + keywords for each transcript.
includeTimestamps	boolean	Return timestamped segments, not just full text.
includeSrt	boolean	Also produce srt and vtt subtitle strings.
language	string	Optional ISO hint (en, es, hi, …). Empty = auto-detect.
proxyConfiguration	object	Apify Proxy (on by default). Switch to Residential if a site blocks datacenter IPs.

{
  "mediaUrls": [
    "https://www.tiktok.com/@nasa/video/7665075736742530317",
    "https://www.youtube.com/watch?v=VIDEO_ID",
    "https://cdn.example.com/episode-12.mp3"
  ],
  "includeSrt": true,
  "includeSummary": true
}

Output

One row per URL:

{
  "url": "https://www.tiktok.com/@nasa/video/7665075736742530317",
  "platform": "TikTok",
  "title": "Something big just landed on TikTok.",
  "uploader": "nasa",
  "status": "ok",
  "transcript": "Something big just landed ...",
  "language": "english",
  "audio_duration": 25.0,
  "duration_seconds": 25,
  "srt": "1\n00:00:00,000 --> 00:00:04,000\nSomething big just landed ...",
  "summary": "A short NASA clip announcing a new arrival on TikTok.",
  "keywords": ["nasa", "space", "announcement"]
}

Failed, private, or non-media URLs return a row with status: "error" and are not billed.

Pricing (pay per result)

Length-tiered, so long recordings stay fair:

Event	Price	When
media-transcribed	$0.030	per item (includes the first audio-minute)
transcription-minute	$0.020	per additional audio-minute
srt-exported	$0.006	only if subtitle export is on
ai-summary	$0.007	only if AI summary is on

A 1-minute clip is $0.03; a 10-minute video is $0.03 + 9 × $0.02 = $0.21. You pay only for successful transcripts.

Notes

• Only media that has an audio track can be transcribed; silent or image-only content returns an error row.
• Very long media works — audio is extracted to a compact stream before transcription, so there's no small file-size ceiling.
• For sites that block datacenter traffic, select a Residential proxy group.