New Apple-backed AI model can generate sound and speech from silent videos

VSSFlow leverages a creative architecture to generate sounds and speech with a single unified system, with state-of-the-art results.