Answers
FAQ
Common questions about Tojito, local processing, models, and the early product roadmap.
What is Tojito?
Local desktop AI for computer audio. On Linux, the validated path is desktop audio → speech recognition → captions → translation → GNOME overlay. Broader platform work (including Windows) continues.
Does my audio leave my computer?
Core filtering is designed to run locally on your device. You should not need to upload audio to a cloud AI service for the basic profanity-filtering experience.
Do I download AI models from the website?
No. Models are managed inside the desktop application — registry lookup, download, verification, install, update, and removal. When a public app build is published, you get the application from the download page; you never fetch model files from this site.
Which languages are supported today?
Source captions and EN↔RU translation are validated on the Linux path. Profanity Filter ON is not verified on the real Pack ASR path.
Is transcription or translation available?
On the Linux MVP path, yes: source captions and offline translation were acceptance-tested with a GNOME overlay. Not a polished multi-platform product release; public download artifacts may still be unpublished.
Which platforms are supported?
Linux desktop MVP is the current validated target. Windows / WASAPI is next and not available yet. macOS remains later. The download page remains source of truth for published artifacts.
How do I send feedback?
Use the contact page or emailhello@dpsonic.com.