Top Privacy and Security

[nuttyartist]

Right now the only information that goes out of users' computers is for performing speech-to-text and text-to-speech using Google services. We should do both offline.

Picovoice already released a superb model to do speech-to-text offline (https://github.com/Picovoice/cheetah), but currently, it only supports Linux. I'll try to contact them for covering more operating systems.

What about text-to-speech?