The Research Scientist will develop new Latent Space Models for voice AI to address data, scale, and cost challenges, focusing on neural codecs and generative modeling for diverse human speech synthesis and understanding.