Vendor page
Cadenza: Open-source voice input for macOS
Cadenza (DragonKingIO, GitHub Pages), 2026
Read the originalRead on October 22, 2026
- What it is
- The maker's home page for Cadenza, an open-source voice input app for macOS. It describes hold-to-speak dictation, local recognition models, vocabulary, and optional AI polishing through a cloud service that the user sets up.
- How much weight it carries
- Written by the maker, so it states intended features in its own terms. The privacy claims are repeated on the maker's privacy page (a separate source). The page names version 1.3.1 and gives no price or plan.
Facts taken from this source
- When AI polishing uses a cloud service, that service receives the recognized text but not the recording.
“A cloud AI service receives the text for polishing, not the recording.” (Feature list)
- Cadenza offers optional AI polishing of recognized text, and lets the user view the original recognition after processing.
“Optional AI polishing View the original recognition after processing.” (Feature list)
- Cadenza says its recognition runs locally, not on someone else's server.
“Recognition runs locally, not on someone else's server.” (Hero section)
- Cadenza lets the user add a vocabulary to correct names, terms and common phrases.
“Use your vocabulary to correct names, terms, and common phrases.” (Feature list, Local cleanup and vocabulary)