Just found this interesting article in my RSS reader (yes I’m slow with catching up my ‘news’ feeds):
It’s a quite good overview of the available components in an open source voice stack. They also cover Rhasspy.
With some interesting broader discussion at the end, for instance under “What’s next for open source voice”:
We’re going to see more software that’s free for personal use but requires a commercial license for enterprise, as Rasa and Picovoice do today. And it’s understandable; dealing with voice in an era of machine learning is data intensive, a poor fit for the open source model of volunteer development. Instead, companies are driven to commercialize by monetizing a centralized “platform as a service.”
And under “What we still need”:
What’s lacking across all these open source tools are user-friendly interfaces to capture recordings and train models. Open source products must continue to improve their UIs to attract both developer and user communities; failure to do so will see more widespread adoption of proprietary and “freemium” tools.
And I’m very sympathetic to the conclusion:
The future of open source rests in your hands! Experiment and provide feedback, issues, pull requests, data, ideas, and bugs. With your help, open source can continue to have a strong voice.
I have always been an avid open source user (20 years now), but it’s only in the last few years that I have been actively contributing to open source projects I use and care about. I have opened more issues on GitHub in the last year than all the time before
Jasper was the first voice assistant project I played with That was three years ago, and even then it was not really maintained anymore, if I remember correctly. I’m also surprised the name keeps popping up everywhere, but many people don’t seem to mind using unmaintained software
It was quite easy to set up, though, so “Beginner” looks like the right qualification to me. But I never got further with it than doing the basics that were included. When I started delving into it to add custom functionality, I found some horrible code and quite rigid architecture. I guess that’s why @synesthesiam started with Rhasspy
Me too. Compiling OpenFST and Phonetisaurus (2013) was a nightmare. I don’t see how Jasper is for beginners while Rhasspy (which has a Docker image) is only for “Experienced Developers”
Yeah, that’s what happened to me. I wanted to link it up to Home Assistant for a project, and found it hard to modify. When I realized they were just doing a “bag of words” speech to text, I started work on Rhasspy using a proper language model.
