Nvidia Unveils Fugatto AI Model with 2.5 Billion Parameters for Audio Creation

Nov 25, 2024, 04:46 PM

Nvidia has unveiled Fugatto, a groundbreaking generative AI model designed for audio creation and transformation. Fugatto, also known as Foundational Generative Audio Transformer Opus 1 and powered by 2.5 billion parameters, can generate and modify music, voices, and sounds using text and audio prompts. Trained on open-source data, the model allows for unique capabilities such as converting piano notes into vocal tracks, altering the emotional tone of voices, and creating entirely new soundscapes. It is aimed at industries like music, gaming, film, and advertising, offering tools for sound design, voice modification, and music composition. Fugatto's ability to produce novel sounds and its potential to revolutionize audio creativity have positioned it as a significant technological advancement in sound engineering.