Blending Udio's AI music with the AI art of Midjourney
In recent months, a wave of powerful AI tools has emerged, giving creators unprecedented capabilities to generate synthetic media. One such tool, Udio, has made waves with its ability to produce impressively human-like vocal performances in the style of famous artists.
I decided to put Udio AI to the test by attempting to create an AI-generated Taylor Swift song.
The process began by placing a prompt in Udio AI:
Prompt: Taylor Swift
Udio replaces it with:
Prompt: Female vocalist, Melodic, Love, Singer-songwriter, Breakup, Romantic, Sentimental, Folk pop, Folk
Rather than directly imitating her voice, Udio instead took contextual cues, capturing the melodic, romantic folk-pop essence that defines Swift's singer-songwriter style.
After a few tries, the AI returned compelling an initial song with lyrics that set the wistful, introspective tone:
[Pre-Chorus]
So I tried to say goodbye
But my heart just keeps on asking
[Chorus]
What if I just stay? (What if I just stay?)
What if time's on our side? (What if time's on our side?)
What if it's not just dreams, can we really fly?
What if I just stay? (What if I just stay?)
Making the Music Video
To accompany Udio's synthetically generated vocals, I turned to AI image generation models like Midjourney, Leonardo AI, and Ideogram.
Since AI tools would struggle to make regular images of an artist, it was necessary to find alternatives and/or aesthetics:
Prompt: Close up of a giant marble statue of Taylor Swift, in the style of Gian Lorenzo Bernini, experimental videos, light azure tones, classical, historical genre scenes, mind-bending murals, dark black background, hyper realism
The resulting AI music video blends Udio's emotionally resonant song with surreal, grandiose Midjourney imagery, achieving an ethereal, dreamlike aesthetic that feels simultaneously modern and classical.
While not directly copying Swift (I hope), the video captures the romantic spirit and introspective storytelling that defines her artistry through an entirely AI-generated lens.
As AI tools grow more sophisticated, creative projects like this showcase their potential to augment and inspire human artists.
While the technology is still emerging, efforts to utilize multiple AI models in innovative ways hint at the imaginative AI-powered experiences that may one day become commonplace.