Audio to Text

Supported files & exports

Upload MP3, WAV, M4A, FLAC, AAC, OGG, WMA, AIFF, AIF, MP4, MOV, AVI or MKV. Files must contain a decodable audio track. This page accepts one local file at a time, up to 192,095 bytes (tested). Cloud links and microphone recording are not supported.

Export TXT, DOCX or PDF documents, or timestamped SRT and VTT subtitles. MP4 and MOV exports burn subtitles into the original video. Audio-only files become captioned videos with a plain background and the original sound. Video rendering takes additional time.

How to transcribe audio to text

Choose a recording, select its spoken language or use automatic detection, then select Transcribe audio. Edit any transcript segment before copying or downloading. Subtitle timestamps come from the same transcription.

Accuracy, privacy & limits

A browser-run Whisper tiny model produces the audio to text transcription. It can make mistakes, especially with noise, overlapping speakers, names or long silences. Review the words and subtitle timing before publishing. There is no speaker identification or translation.

Processing runs inside your browser using WebAssembly. Long recordings can take longer than their playback time and may exhaust device memory. No daily quota or paid transcription API is configured. Audio is held in browser memory during processing; transcripts stay in this tab and disappear on refresh. Only model resources may be cached by the browser. PDF exports use a built-in Latin font; for other scripts, choose DOCX or TXT.