StyleTTS 2: Towards Human-Level Text-to-Speech through Style Diffusion and Adversarial Training with Large Speech Language Models
Cong Han, Yinghao Aaron Li, Nima Mesgarani, Gavin Mischler, Vinay Raghavan
Topics & Concepts
Adversarial systemTraining (meteorology)Computer scienceStyle (visual arts)Artificial intelligenceLanguage modelLinguisticsNatural language processingSpeech recognitionKey (lock)PsychologyPerspective (graphical)DiffusionNatural languageAction (physics)Term (time)Field (mathematics)Topic ModelingSpeech Recognition and SynthesisAuthorship Attribution and Profiling