Do you think it is possible to do network blending with TTS model

[kikirizki]

Hi thank you for your great work. I think your idea about network blending is awesome, but I wonder if it is will work too in another domain like text to speech for example ? I need your personal opinion, however I will try it anyway