Transcription
Choose the spoken language
Spoken language
The first time, a speech model of about 80 MB downloads once and is kept in your browser.
Background noise, music and several people talking at once reduce accuracy. Read the text before using it.
Add a recording
Pick a voice memo, interview, lecture or video. MP3, M4A, WAV, OGG, WebM, MP4 and MOV files work in most browsers.
Choose the spoken language
Pick Turkish, English or Spanish. Choosing the right language gives clearly better results than letting it guess.
Start the transcription
The first time, the speech model (about 80 MB) downloads once and stays in your browser. Then the speech is written out on your device.
Copy or download
Copy the plain text, turn on time stamps, or download a TXT file or SRT/VTT subtitles.
Is my recording uploaded?
No. The audio is decoded and transcribed in your browser. Only the speech model file is downloaded the first time; your recording is not part of that request.
How accurate is it?
Clear speech with little background noise comes out well. Several people talking at once, music, strong accents, names and technical terms cause mistakes, so read the text before you use it.
How long does it take?
It depends on your device. On a typical laptop a minute of speech takes from a few seconds to under a minute; phones are slower. Long recordings can take a while — keep the tab open.
Which files can I use?
Anything your browser can play: usually MP3, M4A, WAV, OGG, Opus, WebM, MP4 and MOV. If a file will not open, convert it with the Video to Audio tool first.
Can I make subtitles?
Yes. Download SRT or VTT; each line carries its start and end time. To burn captions into a video, use the automatic captions in the video tools.
What should I do before sharing the text with an AI?
Before pasting a transcript into an AI chat, you can hide names, numbers and e-mails with the AI Text Cleaner.