Voice Assistants

Do any of you have a foss voice assistant setup at your home? If so what did you use and how did you set it up? I've been looking through the options and I'm very disappointed in what I've found so far. mycroft demands I create an accound with them. Jasper looked the most promising but the packages for my distro were out of date and the build instructions are vauge and specific for debian or raspbian. Same with kalliope and snowboy. I'm just now looking at blather/openassistant. It's less popular and documentation is minimal but it seems like it's just a smaller project.

What exactly do you mean by voice assistant? Something like voice commands? You can always program your own thing using CMU Sphinx, which is a foss library for voice recognition. It's the longest road but it doesn't get any more personal and secure than that.

>>6970 I meant voice command, but I wanted to see how advanced and fancy I could get. Level 1: voice command Level 2: voice response. Level 3: dialog. Most of the projects I was looking at were doing fancy soykaf with neural networks claiming to be AI for some reason. I'd rather keep it metal for now. Ideally I would just have a simple front end like blather seems to be but even it isn't working OBO. I guess I might have run my own pocketsphinx code. But since most of the stuff I was trying was using that as TTS I epxect it might be the culprit. Seeing how the aur package is 5prealpha-9.

This is a blog post that describes how Zuck build his own voice assistant using FOSS software. https://www.facebook.com/notes/mark-zuckerberg/building-jarvis/10154361492931634 Warning: link goes to Facebook.