Audio to Text — Free Transcription

Yes — transcription itself is 100% free and unlimited, with no file count limits, no monthly caps and no watermarks, and all six download formats (SRT, VTT, ASS, TXT, DOCX, PDF) are free too. We run Whisper AI on our own servers so there's no per-minute API cost passed to you. The one optional extra is the AI insights panel — a summary, action items and decisions — which spends one gig credit, and every signed-in account gets 5 free each month.

MP3, M4A, WAV, OGG, FLAC, OPUS, WMA, AIFF — and even video files like MP4, MKV, MOV and WEBM (the audio track is extracted automatically). Maximum file size is 500 MB.

We use OpenAI's Whisper base model which achieves state-of-the-art accuracy on clear speech in English and 90+ other languages. Background noise and heavy accents may reduce accuracy slightly.

Whisper auto-detects the language and supports over 90 languages including English, Spanish, French, German, Hindi, Arabic, Chinese, Japanese, and many more.

Your file is processed entirely on our server — never sent to OpenAI or any third party. Your upload is deleted the moment transcription finishes, and the transcript itself is cleared automatically within an hour.

The transcript is split into chapters every 3 minutes with a timestamp, making it easy to navigate long recordings like podcasts, lectures, or interviews.