Milcer Scribe
Milcer Scribe is a free toolkit that turns audio recordings, videos and text in photos into text. Speech recognition uses OpenAI's Whisper and image text recognition uses Tesseract, entirely in your browser.
MP3, WAV, M4A, OGG, FLAC, WMA, AMR, MP4, MOV, MKV, AVI and every other format are supported. Edit the result and download Word, PDF, TXT or SRT/VTT subtitles.
Everything runs in your browser; your files are never uploaded. The speech model is downloaded once on first use and loads quickly from your device afterwards.
How to use Milcer Scribe
- Choose your filePick an audio, video or image file, or drag and drop it.
- Let it transcribeSpeech or text in the image is converted on your device; you see the progress live.
- Edit and downloadFix the text and download it as Word, PDF, TXT or subtitles.
Frequently asked questions
Is audio to text free?
Yes. There is no time or file limit and no sign-up.
Are my files uploaded?
No. Audio, video and images are processed in your browser; no file is sent to a server.
How long does a long recording take?
On a computer the fast model usually finishes several times faster than the recording's length; phones can be slower. The model is downloaded once (recommended ≈165 MB, fast ≈51 MB).
Which languages are supported?
More than 90 spoken languages; for images, English, Turkish and other common languages.