Turn speech into text and subtitles, on your device
Transcribe interviews, meetings, lectures and videos, then edit while it plays and save the text or SRT and VTT subtitles. The speech model runs in this browser: your recording is never uploaded.
Model
Never uploadedWhisper runs in your browser. Interviews, therapy notes and legal calls stay on your device.
Edit as it playsTap any line to hear it, fix a word, find and replace names, split and join lines.
Subtitles readySRT and VTT for YouTube, Vimeo and editors; captions burned into the video with Pro.
Any lengthLong recordings are cut at quiet moments and transcribed piece by piece as you watch.
How to transcribe audio to text for free
Open a recording or a video, pick its language (or leave Auto), and press Transcribe. The first time, the speech model downloads once (about 77 MB for Fast); after that it opens straight away, even offline. Text appears as each part is done, with timestamps.
Then tap a line to play it and correct anything the model misheard. Find and replace fixes a name everywhere at once. Save the transcript as text or Markdown for notes, or as SRT or VTT subtitles for YouTube, Vimeo, Premiere, Final Cut or DaVinci Resolve.
Transcribe uses OpenAI’s open-source Whisper model, which knows about 100 languages. It works best on clear speech; music, crosstalk and strong background noise make more mistakes, so always read it through. It doesn’t yet tell speakers apart.
Getting ready…
Transcribe
Audio and video to text and subtitles · Free productivity app for your browser
Turn interviews, meetings, lectures, voice notes and videos into text and subtitles, with the speech model running on your own device so the recording is never uploaded. Open a file or record, pick the language or let it detect, and watch the text appear with timestamps. Then edit while it plays: tap a line to hear it, fix words, find and replace a name everywhere, split and join lines. Save plain text, text with times, Markdown, or SRT and VTT subtitles for YouTube, Vimeo and video editors, or open an SRT to edit it. Pro adds the Accurate model, captions burned into your videos for Reels, TikTok and Shorts, translation into English, a word list for names, and unlimited saved transcripts. Uses OpenAI’s open-source Whisper, about 100 languages.
Speech to text on your device: nothing is uploaded
About 100 languages, detected for you
Edit while it plays, find and replace, split and join
Text, Markdown, SRT and VTT subtitles
Record, or open audio and video of any length
Accurate model, burned-in captions and translation with Pro
Works offline
Private: stays on your device
No account or sign-up
Free to use. Optional one-time upgrade: Transcribe Pro, $2.99 (no subscription), or every upgrade in every Yuzool app with Yuzool Plus, $19.99 once. Works on iPhone, Android, Mac and PC; add it to your Home Screen to open it like an app.