Your audio, as text — with every word timestamped
Drop in a recording. AudioToText transcribes the speech — meetings, interviews, podcasts, lectures, voice memos — and gives you a readable transcript with a timestamped timeline, ready to download as TXT, SRT subtitles, WebVTT or JSON.
Drop an audio file here, or click to choose
MP3, WAV, M4A, OGG, FLAC, WebM, AAC · loading your plan…
Please note: recordings often contain other people's voices. Only upload audio you have the right to share — uploads and results are deleted automatically after 72 hours, and you can delete a result immediately from the report below.
What you get from one upload
Timestamps on every line
Each segment of the transcript carries its start and end time, so you can jump straight to the moment in the recording — and the same timings drive the SRT and VTT subtitle files
Four formats, every plan
Plain text for reading, SRT and WebVTT for captions, JSON with segments for your own tools — all four downloads are included on the Free plan, not locked behind a paywall
Charged by the minute
You pay for audio minutes, not files: a 12-minute meeting costs 12 minutes, rounded up to the whole minute. Failed runs never count, and unused minutes reset each month
Long recordings welcome
Free handles files up to 15 minutes, Starter up to 60 minutes, Pro up to 3 hours — upload a whole lecture or podcast episode in one go
Your audio stays yours
Recordings are processed on our own server, never used to train models, and deleted automatically after 72 hours — or immediately, with one click
Email when it is ready
Long audio takes a few minutes to transcribe. Leave your email and we will send the transcript when it is done — or just keep the page open and watch it finish
One recording, four formats
The transcript is plain text you can read, search and paste anywhere. The SRT and WebVTT files are ready-made subtitles for your video. The JSON file carries every segment with its start and end time, so your own tools can build on it — searchable meeting notes, podcast show notes, quote-finding, or captions in your own player.
Questions
What does AudioToText do with my audio file?
It transcribes the speech into written text with timestamps. You get the full transcript plus a segment-by-segment timeline, and you can download it as plain text (TXT), subtitles (SRT or WebVTT), or structured data (JSON with start/end times per segment). Uploads and results are deleted automatically after 72 hours.
What audio formats can I upload?
MP3, WAV, M4A, OGG, FLAC, WebM and AAC. One file per run. If your recording is a video (MP4, MOV), extract the audio track first — for example with ffmpeg: ffmpeg -i input.mp4 -vn -ac 1 -ar 16000 audio.mp3 — then upload the audio file.
How are minutes counted?
By the real length of the audio, rounded up to whole minutes: a 12-minute 20-second meeting costs 13 minutes. The Free plan includes 60 minutes a month, Starter 10 hours (600 minutes), Pro 30 hours (1,800 minutes). The counter resets on the first of each calendar month and unused minutes do not roll over. A run that fails — a corrupt file, or a fault on our side — never counts.
How accurate is the transcript?
Very good on clear, single-speaker audio recorded close to the microphone, and good on ordinary meetings — but automatic speech recognition is never word-perfect. Heavy accents, overlapping speakers, background noise and quiet audio all cost accuracy. Always read the transcript through before you publish it or quote from it, and use the timestamps to check any passage that matters against the original recording.
How long does transcription take?
Roughly a fraction of the recording length for short files, and a few minutes for a long lecture or podcast episode. Uploads run in the background: you can leave the page open and watch the result appear, or leave your email and we will send the transcript when it is ready.
Can I use the transcript as subtitles?
Yes — download the SRT or VTT file and upload it with your video on YouTube, Vimeo or your own player. The subtitle timings come from the same segments shown on the result page, so what you read is what you hear at that time.
Why do I need an account? Transcription works without one.
You do not need one to try: guests transcribe one file at a time — up to 60 minutes a month, files up to 25 MB and 15 minutes — and get all four download formats. An account adds history and the rest of the allowance, and anything you transcribed as a guest in this browser moves into your account when you register.
Do you keep my recordings?
No longer than needed. Uploads and transcripts are deleted automatically after 72 hours, and you can delete any result immediately from the result page or your account page. We never use your recordings to train models, and we never publish, index or share them.