It seems Nvidia “Fugatto” tell us it’s in the music business. Fugatto is an AI-powered sound editor, developed by Nvidia, with the ability to generate music, sounds, and even speech using text and audio prompts it’s never encountered before.
Imagine a Trumpet That Meows
Crazy right? Fugatto can create sounds that are truly incredible or just plain weird. Using text prompts a user can generate sounds like a trumpet that meows, a saxophone that howls, or even “deep, rumbling bass pulses paired with intermittent, high-pitched digital chirps, like the sound of a massive sentient machine waking up.” It’s like something straight out of a sci-fi movie, but it’s real, and it’s here.
But, Fugatto isn’t just about creating bizarre soundscapes – it’s a versatile tool that can handle a wide range of audio tasks. It can generate music based on wild prompts, transform voices by changing accents or emotions, and even edit existing songs by isolating vocals, adding instruments, or swapping out melodies.
To create this audio wizardry, Nvidia researchers fed Fugatto a massive dataset of millions of audio samples, including a library of sound effects from the BBC. They then developed a clever system of instructions that expanded the model’s capabilities, allowing it to perform tasks it wasn’t explicitly trained on.
Fugatto joins a growing chorus of AI audio tools, including those from Stability AI, OpenAI, Google DeepMind, and Adobe. But what sets Fugatto apart is its ability to create entirely new and unheard-of sounds, pushing the boundaries of audio creativity.
Check out this video on how it works.
