The Deepfake Music Industry is Coming
Generative AI will transform Creative Industries, or will they?
Stable Diffusion is pretty dominant. Many artists and designers are using a newly released open source image synthesis model the recent explosion in AI-generated art, commonly known as Stable Diffusion Art (SDA).
Harmony in the Deepfake Metaverse
Harmonai is an organization with financial backing from Stability AI, the London-based startup behind Stable Diffusion. In late September, Harmonai released Dance Diffusion (GitHub), an algorithm and set of tools that can generate clips of music by training on hundreds of hours of existing songs.
One YouTube had a headline that caught my eye:
I made a music video in Stable Diffusion AI... and it's amazing!
Here Olivio is just talking about the video.
He made the Music Video in Stable Diffusion AI with the Deforum SD Google Colab.
He used Video Input with Video from Pexels, that I cut in Davinci Resolve to fit the music first and then rendered it Scene by Scene in SD Deforum Google Colab. He used the same prompt for most of the Video.
Just in one Case he described the Scene to get a better Result. In Davinci Resolve he also added some zoom and movement to most shots to make the a bit more animated. The fading to Black and White and fading out effect is also done in Davinci Resolve and well as the rotation of the video in the second but last scene.
If you are into music, why would you not experiment with Generative-AI’s possibilities?
There is also JukeBox by OpenAI. https://openai.com/blog/jukebox/
As we have seen with text-to-image and text-to-video, text-to-music is also part of this triad. But where does it all lead on the synthetic and deepfake internet of the late 2020s?
Dance Diffusion (taking Cyber-punk to a whole new level) remains in the testing stages -- at present, the system can only generate clips a few seconds long. But the early results provide a tantalizing glimpse at what could be the future of music creation, while at the same time raising questions about the potential impact on artists.
While some artists and designers are embracing generative A.I. it’s disturbing to others. With Harmonai, a cleverly called new method for, you guessed it, creating an audio out of nothing, StabilityAI’s researchers have made (sign up for waiting list for beta) a big advancement in the field of AI audio-generation. The outcomes are astounding and diverse, and they are all just a little bit unsettling.
Even community driven and open-source generative A.I. in music could have very unfortunate consequences. Harmonai is a community-driven organization that publishes open-source generative audio tools to increase everyone’s access to and enjoyment from music composition.
To put it simply, it generates never-before-heard sounds in a process called diffusion, in which random white noise is refined into sounds based on a pre-trained model that contains millions of parameters that were calculated during its training process to encapsulate the characteristics of the source material it was trained on.
To read the entire article consider upgrading to premium.
Keep reading with a 7-day free trial
Subscribe to AI Supremacy to keep reading this post and get 7 days of free access to the full post archives.