Mimic VS Rhasspy

I know it's old at this point and doesn't use the fancy new tech, but Mycroft's Mimic 3 is still pretty impressive and is small enough to fit comfortably and generate speech in real time on a raspberry pi [0]. Some of their voices are better than others, but the best of them are definitely equal to the examples of WhisperSpeech given here. [0] https://mycroft.ai/mimic-3/. - Source: Hacker News / 4 months ago

So dumb, use Mimic3 for free, and you can also use SMML tags so you can change voices for different characters, and modulate the speech to better fit the book! Source: 12 months ago

You might want to check out Mimic 3 by the Mycroft team. Its also open source and runs offline. https://mycroft.ai/mimic-3/. - Source: Hacker News / about 1 year ago

I started out with mycroft which has mimic 3 build in. But you can run it just stand alone as well and quite easy to set up. https://mycroft.ai/mimic-3/. Source: about 1 year ago

So far the Mimic3 from Mycroft has been the best I've seen. With the hifi-tts_low being my favorite English voice. It also support SSML (Speech Synthesis Markup Language) which can allow for more fine tuning on the output (like volume, changes voices, speach rate, specific Phonemes, etc). Source: over 1 year ago

Rhasspy seems promising and I started to tinker with it, but didn't get to a functional state before I got distracted by something else. Source: 5 months ago

Alternatively you could try using rhasspy under termux. Source: 12 months ago

Rhasspy might have a lot of what you're looking for: Https://rhasspy.readthedocs.io/en/latest/. Source: about 1 year ago

There's many voice control options for HA, both private ones like Rhasspy and the corporate spyware ones that only an idiot would use. Source: over 1 year ago

Rhasspy is amazing and more capable than Mycroft (e.g. Satellite support, something that’s mandatory imo), but even more DIY and beginner-unfriendly. That said, the docs are decent, the community is helpful. Source: over 1 year ago