Reply AI Music Contest hosts its first edition at Kappa FuturFestival [Review]

Championing the use of AI and its ever-evolving technology in music and live performances, the first Reply AI Music Contest landed at this year’s edition of Kappa FuturFestival. 

Born to champion the use of AI in music and showcase how it can allow artists and musicians to unlock a whole new creative process at the intersection of AI and live entertainment. We were there to witness the creativity flow on site at the Reply AI Music Contest, and whilst there also was able to gain insight from two participants, including the winner DeLaurentis. 

French composer DeLaurentis, whose entry ‘Sinvocea‘ saw her claim the trophy, utilises AI to visualise her voice in a captivating audiovisual way, and we sat down with her just after her win to learn more about it. 

First of all, congratulations on winning the contest! How does it feel to win?

“It was a very big surprise for me. I didn’t know it was possible. I am very proud.”

What made you originally want to enter the contest?

“A friend on Instagram sent me the link to the contest a few months ago because I use AI in my work. I entered and forgot about it for a while, until a few months later when they told me that I was a finalist and I could come to Kappa FuturFestival. It’s also crazy because I had just learned before flying into Italy that our Great Great Great Grandfather was from Ciconio [Municipality of Turin, where the festival is hosted]. It’s a full circle moment!“

You’ve been using AI in your music since 2018. What drew you to using AI and its technology?

“At the beginning of my musical journey I learned classical music at the Conservatory, and I discovered the work of Laurie Anderson – the female pioneer of electronic music. I was fascinated by her and seeing her working in her studio in the 70s with samplers, vocoders, synthesizers, and with her voice and the violin. That was a starting point to switch the classical way to perform and create music with that technology. And now, technology, AI, and digital tools are my main instruments.”

To talk more about your performance ‘Sinvocea’, can you describe it for us?

“Sinvocea is actually Italian. The name is a mix between synesthesia and voice, vocea because I am a synesthete. Since I was a child I have seen colours and shapes when I listen to music, which is the concept for an album I released a few months ago. 

I made a correspondent board between colours, instruments and emotions, and the album is full of red, green, purple and orange. For my performance at the Reply AI Music Contest I performed the red, green, purple and orange. Sinvocea is an AI created in collaboration with Sony CSN. I have used their tools for production for years, but one year ago they asked me: what is the AI of your dream? I told them that I would like to interact in real time on stage using the AI to improvise. 
At the beginning I would sing something and the AI would answer me, but because of my synesthesia and the concept of the album, we decided to develop the AI with visuals. It’s a conversation between my voice and the visuals.”

How does it technically work?

“Its model was training with the pitch of my voice but also the volume, vibrato and breath. The AI starts from nothing, just analyses the audio of my voice and with that it generates in real time visuals. It’s never the same because it’s the visuals of the moment.”

How long was the performance that you presented for the Reply AI Music Contest in production for? 

“I think for this performance and this album I worked for two months just to create the show. With this stage setup I used the minimum, as I have to plug my equipment in, but when I have the time for that, I have many digital instruments and they are very sensitive. It takes a long time to configure. At the contest, the challenge was there due to the heat, but I felt as though if I could perform today, I could perform anywhere and any festival in the world. It was also my first time performing in the booth like a DJ.”

Another entrant who beat out hundreds of applications to bring his concept to the Reply AI Music Contest live was I AM JAS, who alongside his VJ The Glad Scientist, presented his handcrafted audio visual live show. Sitting down with the pair, we got another insight into the contest directly from the source.

How does it feel to have been able to bring your performance as a finalist to the contest?

I AM JAS: “It feels amazing. It was always my dream to play at Kappa FuturFestival, but playing with my live show, so not as a DJ is something that I didn’t expect. Being here with my mentor [The Glad Scientist] and feeling his support has been amazing.”

How did you two connect?

The Glad Scientist: “I was his mentor and teacher last year in a master’s program in Barcelona, so I was teaching him a lot of AI techniques for, like, company UI, and also teaching touch designer more on the visual side of stuff. But he came in with a very good background from the audio side, as someone who, as a musician, was like, Hey, I see that live shows are going more into this audio visual space, I want to learn how to do all the visual things so that I can start like performing more as an audio visual artist. I do a bit of everything audio and visual, but visuals is what I’m teaching.”

How did you originally connect with AI and its use in music?

I AM JAS: “I use AI as as a tool. When I’m stuck in some in some song, and I’m not able to get out that sound that I see clear in my in my mind, I use AI to inspire myself. And also I’m learning a lot even now, also in the master programme last year, and learning about AI for making visuals. I have always been an electronic music artist, and as a DJ or as a producer, having the possibility to create stories that I that I imagine when I’m making the songs and put it into a visual format is something that is simply amazing.”

The Glad Scientist: “In terms of AI in visuals, I love it and embrace it a lot. Using it as a tool and in a larger sort of workflow or vision usually is my preference, and because I’ve seen enough of people just using it as their art, that I’m like, All right, well, there’s more out there. You can do more with it. I can respect different techniques, of course. If someone’s using Midjourney and Runway, or a lot of custom workflows, I can get behind it. You can see that they have a vision. They’re really trying to express something.”

View this post on Instagram

A post shared by I AM JAS (@i_am_jas_music)

The I AM JAS performance is inspired by Antoni Gaudí’s mosaic technique. Can you tell us a bit more?

I AM JAS: “In electronic music, we use the technique of sampling, and Antoni Gaudí also did this trencadís technique which is a mosaic technique in which you use certain small pieces of ceramic or glass, and with that, he made a new composition. I visualised that relation between something that is not connected at the beginning, that you can then mix.”

There are a lot of elements that went into your performance at the Reply AI Music Contest. Can you tell us about it?

I AM JAS: “I’m a geek, and my brain doesn’t stop. I like to learn and I like to create, and I just find the solution is to connect everything and to put more more story in the music. I use a lot of tools to make it possible. I’m an engineer, so I love the challenges!”

The Glad Scientist: “My work usually has a different workflow for every single performance and piece that I work on, and I was working for a long time in a virtual reality headset, like modular synthesis and controlling audio visual sets that way. So my work was pretty different in general. But for this one, it was really cool, because Jas had all these like techniques that he was mixing, including Antoni Gaudí’s, and like, creating these custom models and then using them in ComfyUI to train on architecture from Gaudí, and also train on like landscapes from his own town. He was making the workflows. I was just fixing errors, to be honest. The live set is all connected through Ableton and touch designer, so that it’s controlling everything as he’s performing. It’s really reactive.”

I AM JAS: “I prepared the entire Ableton project. For my music, my own songs, I was using a microphone tool to sing. The Glad Scientist is using resolute, and he’s receiving messages from my music.”

The Glad Scientist: “Jas asked me if I wanted to help adapt the visuals, because originally, it was made for an immersive space at a very specific resolution. And then here at Kappa, it’s a box, so it’s like the reverse of an immersive space. You’re, like, around the screen instead of inside of it. So when he asked me to help him adapt his visuals, I offered to VJ for him too. Then as we met to rehearse, we were talking and started to adapt the resolution of stuff, and then also connected Touch Designer to Ableton so that his set could just be really integrated with controlling the visuals.

At that point, it was like a moment where we both stopped and we’re like, do you want to make some new content for this? Then, in the last two and a half weeks leading up to the contest, we just kind of went absolutely crazy making completely new content for five out of the six songs.”

Proving to be more than a success, the future of the contest is extremely bright and we can’t wait to see where it goes next. To find out some more information about what went down this year, visit their Instagram page here and their website here.