开源中国·· 22 days agoAI score63
Google DeepMind releases Genie 4, world model begins generating sound in sync
Genie 4 来了,谷歌 DeepMind 的世界模型开始同步生成声音
AI briefing
Google DeepMind releases Genie 4, world model achieves native multimodality, with visuals, music, ambient sound, and dialogue generated simultaneously. Technically based on the Gemini family, Gemini Omni is responsible for anchoring the simulation to the appearance and physical behavior of the real world.
Open original oschina.net
Google / VeoVideo GenerationDubbing and AudioModel Releases#Model release#Video generation#Voice & audio#Google