Is this audio to text converter free?

Yes. It is free with no sign-up and no per-minute charge. Transcription runs on your own device in your browser.

Is my audio uploaded anywhere?

No. Your file is decoded and transcribed inside your browser tab. The only download is a one-time copy of the compact Whisper speech model (about 40 MB) from a public model library, which your browser then caches.

Which audio formats can I use?

MP3, WAV, M4A and MP4 work well, plus any audio or video format your browser itself can play. If a file cannot be decoded, the tool says so plainly instead of guessing.

How accurate is the transcript?

It is at its best with short, clear recordings: one speaker, little background noise, roughly a few minutes at a time. Long, noisy or multi-speaker recordings will contain more mistakes, and names and technical words are usually the first to go wrong. Always proofread before relying on the text.

Can I get subtitles for a video?

Yes. After transcribing, download the .srt or .vtt export — both carry the model's timestamps as subtitle cues. If you edit the transcript text, the subtitle export falls back to one cue holding the edited text.

Does it work offline?

After the first visit, the speech model is cached by your browser, so transcription keeps working without an internet connection. The very first transcription needs to be online to fetch the model.

Расшифровка аудио в текст онлайн — транскрибация