synesthesiam

On my list to check out tomorrow :D

As the sibling comment mentions, the next version of Piper will no longer use espeak-ng to avoid potential GPL licensing issues.

Yes, it is possible by installing the Wyoming Satellite software on the Mark II. All the pieces are there, it's just missing some easy-to-install firmware. Source: I worked at Mycroft on the Mark II and am now the voice guy for Nabu Casa (Home Assistant).

Look into training a Dreambooth model. Huggingface has a guide for this.

Thank you for the kind words, @follower!

I'm the author of Piper; it is a successor to Larynx (originally named Larynx 2). Piper uses the same underlying model as Mimic 3, which I developed before joining Mycroft. However, Piper uses a different library to get word pronunciations, so the voices aren't compatible between the two projects.

It's been an awesome year so far with Nabu Casa, and I'm very fortunate to be able to work on something I love. I hope to contribute to the open source voice space for many years to come :)

Coauthor of the blog post here. You're right, I said it on the live stream we had today but forgot to mention it in the blog post: the i5 is from a Lenovo ThinkCentre M72e. They're available refurbished for less than the cost of a Pi 4 these days, so it seemed to be a good comparison!

Home Assistant and Rhasspy 3 will be able to share voice services thanks to the shared Wyoming protocol. Rhasspy 3 will have more options, including lots of experimental services.

Mark I or Mark II?

This has always been a struggle. Rhasspy can gather lists of songs, artists, etc. but it will have to guess many of their pronunciations. And it seems artist/band names often purposely thwart conventional pronunciation rules :P

They share a lot of the same pieces, but voice2json is meant to work in Unix-style pipelines. Rhasspy has MQTT/HTTP/Websocket APIs instead.