Google DeepMind Unveils Video-to-Audio Generative Technology, Releases New Paper

Jun 17, 2024, 04:10 PM

Google DeepMind has announced significant progress in its video-to-audio (V2A) generative technology. This new system can add sound to silent video clips that match the acoustics of the scene and accompany on-screen action. The technology uses video pixels and text prompts to sync rich soundtracks with silent videos, potentially revolutionizing filmmaking. Examples provided include audio generated for scenes such as 'Jellyfish pulsating under water, marine life, ocean.' DeepMind claims the V2A tech can generate an unlimited number of tracks for any video. They have also released a new paper on this technology.