Turn recordings, sermons, lessons, meetings and videos into text and subtitles with AI. Supports English, Korean and 90+ other languages.
🧠 AI model
Fast: about 150 MB download. Good for clear English.
🌍 Language
Record
Talk, then press stop
Working…
The text will appear here…
Tip: click a time to play from there. You can click on the text to fix a word before downloading.
🔒 Everything happens inside your browser — your recording is never uploaded. The AI model downloads once and is then saved for next time.
About this free Speech to Text
Turn recordings into text with OpenAI Whisper running right in your browser. Drop an audio or video file — a sermon, lesson, interview, meeting or voice memo — or record with your microphone. The words appear part by part with times, you can fix any word, and you can save the result as text or as SRT and VTT subtitles for YouTube and video editors. It also understands Korean and more than 90 other languages, and can translate speech into English.
How to use it
Choose Fast or Accurate, and the language that is spoken (or let it detect it).
Drop an audio or video file, or press the red button to record.
Watch the text appear with times. Click a time to listen again, or click the words to fix them.
Copy the text, or download TXT, SRT or VTT.
Features
Whisper AI speech recognition
Audio and video files: MP3, M4A, WAV, MP4, WEBM, OGG and more
Record from your microphone
English, Korean and 90+ languages, plus translate to English
Timestamps with click-to-play
Edit the text before saving
Download TXT, SRT and VTT subtitles
Free, no sign-up, never uploaded
Frequently asked questions
Is this speech to text tool free?
Yes. It is free with no sign-up, no minutes limit and no watermark.
Is my recording uploaded?
No. The AI runs inside your web browser, so your audio never leaves your device.
Does it understand Korean?
Yes. Choose Accurate and set the language to Korean for the best results. It also supports Spanish, Chinese, Japanese and many more.
Can I make subtitles?
Yes. Download SRT or VTT files and add them to YouTube, Premiere, DaVinci Resolve, CapCut and other video tools.
How long can the recording be?
Long recordings such as a full sermon work. The text appears part by part; a computer with a graphics card is much faster.
What is the difference between Fast and Accurate?
Fast uses a smaller AI (about 150 MB) that is quick and good for clear English. Accurate uses a larger AI (about 330 MB) that is better for other languages, accents and noisy audio, but slower.