About convert.express

Private, pay-as-you-go audio & video transcription in the browser, powered by OpenAI Whisper, Microsoft MAI-Transcribe 1.5, and Google Gemini 3.5 Transcribe. Upload MP3/WAV/M4A/FLAC/MP4/MOV or paste a file URL — get a transcript in seconds, plus optional summaries, action items and translation. 10 free minutes daily, €0.02/minute after, €0.50 minimum charge, no subscription or account required to try. Files auto-delete after 24 hours and are never used for AI training.

Who builds it

Alexey Babkin, founder: I was looking for a simple, convenient tool for occasional transcription. Everything I found at the time was either subscription-based — too expensive for how little I actually needed it — or functionally overkill, and priced to match. So I built convert.express, for myself and for other people running into the same problem.

GitHub: github.com/droozhok · Quora: quora.com/profile/Droozhok · Product Hunt: convert.express · LinkedIn: linkedin.com/in/babkin

Why it's priced this way

I priced it the way I wished the tools I'd found were priced: 10 free minutes every day, then €0.02 per minute with a €0.50 minimum charge — no subscription, no account required to start. That fits how I actually use it myself — a few interviews a month, an occasional lecture, a one-off meeting recording — and I'd rather say plainly that it's a poor fit for constant, high-volume transcription than pretend otherwise. If you're transcribing hours of audio every day, a flat-rate subscription product will almost certainly cost you less. Full breakdown on pricing.

The free 10 minutes a day covers more than you'd expect — a handful of long voice messages, for instance. I also use convert.express myself to paste in a YouTube link and get a transcript, summary, and action list out of a video, without opening a subscription tool just for that.

Data handling

  • Audio files and transcripts are deleted automatically within 24 hours.
  • Your files are never used to train AI models.
  • Storage uses UUID-based object keys — never your original filename.
  • Transcript text and file contents are never written to logs; only metadata is.

Full details, including what data we collect and why, are in the privacy policy.

What it runs on

Transcription runs on OpenAI Whisper, Microsoft MAI-Transcribe 1.5, and Google Gemini 3.5 Transcribe. Optional summaries, action items, and translation run on Anthropic Claude. Naming the sub-processors that touch your audio is not a leak — it is information you are entitled to before you upload anything.

When not to use convert.express

This is not the right tool for every job. Specifically:

  • High meeting volume. If you need every meeting captured automatically and searchable later, a dedicated meeting bot with a persistent archive will serve you better than a manual upload-per-file workflow.
  • Teams needing seat management. There is no team plan, seat administration, or organization-wide retention policy control here.
  • Material that legally cannot leave your infrastructure. If your data residency or compliance requirements mean audio cannot be sent to a third-party API at all, self-hosting Whisper is the correct answer, not this service.
  • Anything requiring a signed BAA or equivalent. convert.express does not execute Business Associate Agreements or comparable contractual data-processing commitments.
  • Anything you might need to recover later. 24-hour deletion is unconditional. If you don't download your transcript in time, there is no archive to recover it from — that is by design, not a bug.