Ad
Short

AI voice startup ElevenLabs has given a first look at a new model that can generate songs from prompts. The company, founded by former Google and Palantir employees, specializes in machine learning to clone and synthesize voices. Now, ElevenLabs seems to be targeting the music industry with ElevenLabs Music. The published examples are roughly on par with the competition from Suno AI and Udio, and have a running time of around three minutes.

Video: Elevenlabs via X

Ad
Ad
Short

The research team at Temporal Games has developed a plugin for real-time streaming of volumetric video using Gaussian splatting in Unreal Engine 5. Animated Gaussian splatting makes it possible to represent real events or performances as three-dimensional projections of reality. The challenge lies in the large amount of data required to render the animation. The team was able to reduce the data size by a factor of 15. In a recently published video, the researchers also show how the integrated image can be dynamically lit in real time. The team plans to develop a client-server solution for 4DGS streaming and to make the technology available for games and web platforms.

Ad
Ad
Ad
Ad
Google News