MP3, WAV, M4A and video files

Audio to text, with the structure the raw file does not have

Converting a file to text is the easy part. The useful part is what comes next: timestamps, readable paragraphs, translation, subtitles and a summary that tells you what is inside without listening to all of it.

Upload an audio or video file, choose the spoken language and let Whisper Large v3 Turbo handle the transcription. Export the result as plain text or subtitles, or translate it into English, Chinese or Spanish.

Start recording

Common formats

MP3, M4A, WAV, OGG, WEBM and the audio track of MP4 files are supported in the browser.

Subtitles included

The model returns WebVTT, and VoxNote converts it to SRT so you can publish captions immediately.

Translation on demand

Transcribe in the original language, then translate the transcript or summary into your target language.

No desktop software

Everything runs in the browser, including on phones and tablets. There is nothing to install or update.

How it works

1. Choose a file
Drop in an audio or video file up to the size limit, or record a new voice memo.
2. Pick the language
Auto-detect works for most files, but choosing the language improves accuracy for short recordings.
3. Transcribe
The file is split into safe segments and transcribed sequentially with progress feedback.
4. Export or translate
Download TXT, SRT, VTT or JSON, or ask for a translation and a summary.

Frequently asked questions

Can I convert a video to text?

Yes. Upload an MP4 or another supported video file and the audio track is transcribed.

Is there a file size limit?

You can open files up to 25 MB, but one Cloudflare AI request is limited to two minutes or five megabytes. Record inside VoxNote or split long files into shorter segments before transcription.

How long does it take?

Transcription speed depends on audio length and current capacity. Short files usually finish in seconds; long recordings are processed segment by segment.

Can I get a translation without a transcript?

You always get the transcript first. Translation is a separate step so you can verify the original words before using the translated version.