Voice in.
Meaning
out.
From spoken audio to readable text.
Punctuation. Paragraphs. Still your words.
One independent engine. Any server-side integration.
Let's meet at nine. Bring the notes.
We can work through them together.
Not a rewrite.
Illustrated sample · no recording or upload on this page
Structure the speech.
Keep the speaker.
Vox Homini separates speech processing from the app around it. Send audio from your server. Get the transcript and its structured version back.
- AudioYour spoken input
- RecognitionSpeech becomes text
- StructurePunctuation, case, paragraphs
- Your appBoth versions, ready to use
No paraphrasing. Text structuring preserves the spoken words and their meaning. It is not a summary or a new version of your thought.
A service, not a browser secret. Recognition and text structuring use external providers. Your application calls Vox Homini from its server.
An engine.
Not another app.
Trubado is the first integration. Vox Homini is the independent speech layer underneath it — made to be called by other products, too.