Audio Transcription: Speech to Text, Video to Text, Whisper avatar

Audio Transcription: Speech to Text, Video to Text, Whisper

Pricing

Pay per event

Go to Apify Store
Audio Transcription: Speech to Text, Video to Text, Whisper

Audio Transcription: Speech to Text, Video to Text, Whisper

Audio transcription and speech to text with Whisper: MP3, WAV, M4A, MP4, WebM files to text, timed segments and SRT subtitles. Auto language detection, no API key. $0.006 per minute ($0.012 accurate mode); failed links free. Input: {"url":"https://.../talk.mp4"}.

Pricing

Pay per event

Rating

0.0

(0)

Developer

Isaac

Isaac

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

13 days ago

Last modified

Features

Limited permissions

Share

Audio Transcription: Speech to Text, Video to Text & Podcast Transcripts (Whisper)

Turn audio and video into text, timed segments and SRT subtitles. Give it a direct file link, a podcast RSS feed (transcribes the newest episodes) or a media page such as SoundCloud or archive.org, and get a clean transcript back, with the language detected automatically. At $0.006 per minute it is one of the cheapest Whisper transcription APIs on Apify.

Use this when...

  • An agent or workflow needs to read a podcast, interview, meeting recording, lecture or voice note.
  • You need subtitles (SRT) for an MP4, MOV or WebM video.
  • You want to search, summarize or quote spoken content with timestamps.
  • You need speech-to-text without managing GPUs, API keys or OpenAI accounts: pay per minute, nothing else.

Supported: MP3, WAV, M4A, AAC, OGG, OPUS, FLAC, MP4, MOV, WebM, MKV and most other formats ffmpeg can read. About 99 languages (Whisper). Also supported: podcast RSS feeds (set Episodes per podcast feed), and media pages on SoundCloud, archive.org, Bandcamp and the many other public sites that yt-dlp supports. Not supported: YouTube (YouTube blocks cloud servers without paid residential proxies), Spotify-exclusive shows, and anything behind a login. For a podcast that is also on YouTube, use its RSS feed instead; almost every podcast has one.

Use in ChatGPT or Claude

Add this tool to your AI assistant as a custom connector (MCP server) with this URL:

https://mcp.apify.com?tools=poised_beardlichen/audio-video-transcriber
  • ChatGPT: Settings → Apps & Connectors → Create (needs Developer mode), paste the URL, choose OAuth and sign in to Apify.
  • Claude: Settings → Connectors → Add custom connector, paste the URL, then click Connect.

Runs go on your own Apify account at the same per-event prices, so the free plan's monthly credit applies.

Try it free

Apify's free plan includes $5 of usage every month, no credit card. That's about 800 minutes (13+ hours) of audio a month in standard mode, which is enough to test a whole podcast back catalogue before paying.

Pricing (pay per event)

ModeModelPrice
Standard (default)Whisper base$0.006 per audio minute
AccurateWhisper small$0.012 per audio minute

Minutes are rounded up per file. Failed links, web pages and failed downloads are free.

Examples:

  • 5-minute voice note, standard: 5 × $0.006 = $0.03
  • 34-minute podcast episode, standard: 34 × $0.006 = $0.204
  • 1-hour interview, accurate: 60 × $0.012 = $0.72

Control your spend: set Max minutes per file to transcribe only the beginning of each file, and set the run's maximum charge (Apify's "Max cost per run" option). The actor stops before going over it and never charges more than your limit.

Input

{
"url": "https://archive.org/download/ElephantsDream/ed_1024_512kb.mp4",
"model": "base",
"language": "",
"maxMinutes": 0
}
  • url / urls: direct media links, podcast RSS feed URLs or media page links.
  • maxEpisodes: for RSS feeds, how many of the newest episodes to transcribe (default 1).
  • model: base (standard), small (accurate) or tiny (fast, standard price).
  • language: ISO code such as en; blank means auto-detect.
  • maxMinutes: only transcribe the first N minutes of each file (0 = all).

Output

One dataset item per file:

{
"url": "https://.../talk.mp3",
"success": true,
"model": "base",
"language": "en",
"durationSec": 2013.4,
"chargedMinutes": 34,
"text": "Chapter five. ...",
"segments": [{"start": 0.0, "end": 4.2, "text": "Chapter five."}],
"srt": "1\n00:00:00,000 --> 00:00:04,200\nChapter five.\n"
}

Failed files return success: false and an error message, and aren't charged.

Use with AI agents (MCP)

Add this tool to Claude, Cursor or any MCP client:

https://mcp.apify.com/?tools=poised_beardlichen/audio-video-transcriber

The agent passes a media URL and gets the transcript back. You can also call it from the Apify API, the Python or JavaScript client, Make, Zapier or n8n.

FAQ

How long does it take? Standard mode usually runs several times faster than real time; a 30-minute file takes a few minutes. Is my audio stored? Files are downloaded into the run, transcribed and deleted. Only the transcript is saved to your dataset. YouTube links? Not supported: YouTube blocks cloud servers unless paid residential proxies are used, and we'd rather keep the price low. These links fail fast and are free. For podcasts, use the show's RSS feed. How do I transcribe a podcast? Paste the podcast's RSS feed URL (on Apple Podcasts pages it's listed as the feed; or search "

Other tools by us

Support

A file that won't transcribe? Open an issue on the Issues tab with the link.