Audio to Transcript
Upload an audio or video file and get the text back. Files are forwarded to the configured speech-to-text provider and never stored.
File transcription is not switched on yet, so you cannot upload a file here for now. Transcripts from YouTube videos work today and need no file.
An audio file in, a transcript out. This page is for the straightforward case: you have a recording, you need it written down, and you would rather not listen to it again to type it out.
The transcript preserves what was said in order, ready to copy into notes or a document. It is a record of the speech, not a summary of it.
Frequently asked questions
- How accurate is the transcript?
- It depends on the recording. Clear speech from one speaker in a quiet room is close to perfect; heavy accents, crosstalk, and background noise all degrade it.
- Will it punctuate and capitalise?
- Most modern speech recognition adds sentence punctuation and capitalisation. Where it does not, the transcript comes back as unpunctuated running text.
- Can I edit the transcript afterwards?
- Yes — it is text. Copy it into whatever you are working in and correct the names and terms the recogniser got wrong, which is usually a small fraction of the total.
- Does it identify who is speaking?
- No. Speaker labelling is a separate capability and is not part of this transcription.
- Is my audio uploaded?
- Yes, to the configured speech-to-text provider, because recognition cannot happen in the browser. It is not retained by this site.