Training new larynx voice for Rhasspy

Can anyone point me (a newbie) towards a tutorial on customising a new voice from my own voice samples from spoken sentences through to getting it up and running in Rhasspy’s TTS using larynx? I have seen its-prompts, but the steps from that to a new voice is still a mystery.

Also larynx support SSML but I haven’t been able to get the expected sounds when an intent handler outputs SSML as a string. Can anyone enlighten me on how to get that working?

Big thanks for such a wonderful delight that is Rhasspy!

1 Like