Nvidia launches AI audio model Fugatto for music and sound creation

Nvidia has launched a new AI audio model called Fugatto, which can create and modify music and sounds based on text prompts. Users can generate music snippets, alter existing songs, and produce unique sounds. Fugatto stands for Foundational Generative Audio Transformer Opus 1. It allows users to change instruments, accents, and emotions in audio files. Nvidia claims Fugatto is the first model to demonstrate emergent properties, showcasing advanced capabilities from its trained abilities and the ability to follow free-form instructions.