MP3 transcription

MP3 to text, with a transcript you can actually use

Upload a MP3 file for podcast episodes, downloaded interviews and voice recordings and get a timestamped transcript, a summary and subtitle files. Nothing needs to be installed, and you can start with the free daily allowance.

MP3 is already compressed, so accuracy depends on the original bitrate. For speech, 64 kbps mono is usually enough; 128 kbps and above gives the model more detail.

Start recording

Timestamps and segments

Every paragraph keeps its start and end time so you can jump back to the exact moment in the audio.

Summary, quotes and actions

Ask for key quotes, action items or show notes instead of a generic recap.

Subtitles included

Export SRT or WebVTT for YouTube, social video or your podcast host.

Browser-based and private

The file is processed in memory for the request and is not stored in our object storage.

How it works

1. Upload the MP3 file
Drop it into the recorder or record a new voice memo in the browser.
2. Choose the language
Auto-detect works for most files; choosing the language helps with short or noisy recordings.
3. Transcribe
Long recordings are split into short segments and processed with progress feedback.
4. Export
Download TXT, SRT, VTT or JSON, or translate the transcript into English, Chinese or Spanish.

Frequently asked questions

Is MP3 to text really free?

You get a daily free allowance powered by Cloudflare Workers AI. Long recordings or heavy use can be upgraded to Pro minutes.

Can I get subtitles from a MP3 file?

Yes. The transcript includes timestamps and can be exported as SRT or WebVTT.

What about sound quality?

MP3 is already compressed, so accuracy depends on the original bitrate. For speech, 64 kbps mono is usually enough; 128 kbps and above gives the model more detail.

Do you keep the file?

No. The audio is processed for the transcription request and is not written to object storage by VoxNote.