Speak or type, and get a handwritten page
Speak a paragraph, get a handwritten page — for cards, letters, journals and social posts
Processed locally, text never uploadedText (fix recognition mistakes)
0 charsThis browser doesn't support speech recognition. Chrome / Edge recommended, or just type on the right.
Handwriting
FontPaper
More like handwriting
Speak, and it writes itself
Tap "Tap to speak" below, or type in the editor on the left
FAQ
Speech recognition doesn't respond — what should I check?
Use desktop Chrome / Edge, allow microphone access (the icon left of the address bar) and check your system microphone. On unsupported browsers, type directly into the text box.
The transcription has mistakes — can I fix them?
Yes. Edit the text box: the handwriting on the paper updates in real time.
How do I make a handwritten greeting card?
Speak or type your message, pick a font, paper and ink, then export as PNG or PDF — under 3 minutes. The template library gives you ready-made copy for common occasions.
Can I use the generated images commercially?
Yes — generated images belong to you. Fonts are open source under SIL OFL. One limit: don't impersonate another person's handwriting (see Terms).
Is audio-file transcription really free?
Yes. Transcription runs in your browser with an open-source Whisper model — no account, no minute caps. The model downloads once (~80MB) and works offline afterwards.
Which audio formats are supported?
Anything your browser can decode: mp3, wav, m4a, ogg, webm, flac. Safari's decoding is incomplete for some codecs — desktop Chrome / Edge recommended.
Will every export look identical?
You control it: "Shuffle strokes" re-generates the writing-style fingerprint (slant, baseline wave, size rhythm, darkness), realism sets how messy it looks, and your preferences are remembered locally.
Voice to Handwriting · voicetohandwriting.online