“Nvidia demonstrated an AI model for creating music and audio. She can change voices and generate new sounds. Fugatto technology is designed for music, film and video game creators. The neural network is capable of generating sound effects and music upon prompting. For example, you can create “audio trumpet barking like a dog”. Another example cites the sound of “deep, rumbling bass pulses combined with intermittent high-frequency digital chirps—like the sound of a huge intelligent machine, […]”, — write: businessua.com.ua
Nvidia demonstrated an AI model for creating music and audio. She can change voices and generate new sounds.
Technology Fugatto intended for music creators, movies and video games.
The neural network is capable of generating sound effects and music upon prompting. For example, you can create “audio trumpet barking like a dog”. Another example cites the sound of “deep, rumbling bass pulses combined with intermittent high-frequency digital chirping—like the sound of a huge intelligent machine waking up.”
A distinctive feature of Nvidia’s solution is the ability to analyze and modify existing audio. For example, it can turn a melody played on the piano into human singing.
“If we think about synthetic audio over the past 50 years, now music sounds different thanks to computers, synthesizers. I think generative AI will bring new possibilities to music, video games and ordinary people who want to create new things,” commented Brian Catanzaro, vice president of deep learning research at Nvidia.
The company’s new model is based on an open source database. The company is considering options for its presentation to the public.
“Any generative technology always carries some risks, because people can use it to create things that we don’t want them to create,” Catanzaro emphasized.
We will remind you that Google DeepMind announced the development of technology based on artificial intelligence to create soundtracks for videos.
The source