Quoted By:
Wait so according to Vedal the first 40 or so minutes of this stream was Neuro's LLM with Kiara's voice? But how in the world was she able to laugh and move around and make so many different tonal expressions? Is Neuro really constrained by her current tts model this badly?