Nvidia , a company that specializes in manufacturing artificial intelligence chips and software , has announced a new model called “Fugatto” that aims to produce innovative music and sound effects and modify sounds. This innovation is aimed at music, film, and video game makers, as it can generate new sounds or modify existing audio clips in creative ways
Model capabilities
Generate new sounds: It can produce unique sound effects such as turning the horn sound into a dog barking
Edit audio clips: The model can modify existing sounds, such as turning a piano piece into a human-sounding vocal, or changing the tone and mood of a previously recorded voice
In this context, Brian Catanzaro, Vice President of Deep Learning Research at NVIDIA, explained that artificial intelligence will contribute to a qualitative shift in music, video games, and the content industry in general.
“If you look at the evolution of audio over the past 50 years, computers and electronic music devices have changed the way we hear music,” he said. “I think generative AI will bring new capabilities to these industries and to everyday people who want to create.”
Despite the potential, Nvidia has said it has no plans to release the model to the public at this time, citing concerns about misuse of the technology, such as creating content that could infringe on intellectual property rights or be used to spread misinformation. “Any generative technology carries risks, and we have to be careful before launching it,” Catanzaro said
The announcement comes as major tech companies like Meta and OpenAI compete to develop AI models that can generate audio and visual content from text.
However, these technologies face legal and ethical challenges, such as the case of actress Scarlett Johansson accusing OpenAI of copying her voice without permission.
This technology is a new step in the world of digital creativity, but it raises questions about how to balance innovation with protecting intellectual rights and users