Honored 🎷🎸🥁 𝗠𝗶𝘅𝗔𝘀𝘀𝗶𝘀𝘁 🎷🎸🥁 was selected as the #COLM2025 oral spotlight. Go check out @mclemcrew.bsky.social's 𝐭𝐚𝐥𝐤 on 𝐖𝐞𝐝 (𝐎𝐜𝐭 𝟖) at 𝟑:𝟑𝟎𝐩𝐦 in 517BC and 𝐩𝐨𝐬𝐭𝐞𝐫 from 𝟒:𝟑𝟎-𝟓:𝟑𝟎 in 710!

Mega stoked to attend #COLM25 this week and present our work, MixAssist, on Wednesday!

@anamarasovic.bsky.social sadly can't make it 😭, but hit me up if you'd like to chat about audio language models, music mixing, or anything else regarding music and audio!

Beyond excited to share that I’ll be joining Universal Audio as an LLM Applications Engineer!

So stoked to continue working on co-creative AI elements of music production in this role 🎹

S/O to @anamarasovic.bsky.social for teaching me all things NLP these past few years!

My first audio AI paper, thanks to @mclemcrew.bsky.social who introduced me to a whole new world of co-creative music producing! Getting to work with students who bring their unique passion into PhD work is one of the best perks of a professor's job.

Check more in the thread👇🏻 Soon at #colm2025

Check out the paper: arxiv.org/abs/2507.06329

And the website: mclemcrew.github.io/mixassist-we...

S/O to @anamarasovic.bsky.social for all the leadership, support, and expertise they put into this work! We are so happy with how it turned out and hope it helps the community of music producers!

The MixAssist Dataset consists of 431 audio-grounded conversational turns from 7 in-depth sessions, designed to train AI assistants that can provide instructional, context-aware guidance on music production pedagogy.

Check out the full dataset here: huggingface.co/datasets/mcl...

While AI is recontextualizing music production, many tools focus on automation, leaving a gap for artists who want to learn. To address this, we developed MixAssist, a novel dataset capturing the dialogue between expert and amateur producers during collaborative mixing sessions.

Absolutely stoked to announce that our paper “MixAssist: An Audio-Language Dataset for Co-Creative AI Assistance in Music Mixing” was accepted at COLM’25!

This research aims to foster development of AI that empowers artists by augmenting their skills, not automating their flow.

Join us Friday 11am London time for an AIL Open Seminar by Maxime Popoff on low-latency multichannel FPGA audio synthesis. Streaming at youtube.com/c/AugmentedInstrumentsLab or DM for in-person joining details!

Did not expect this at all, but absolutely thrilled to be selected to contribute this summer!

I’ll be working on Tölvera and making the codebase a bit more accessible to non-technical audiences through natural language interfaces :)

“DJs aren’t real musicians, they just push buttons and music comes out”

Pianists:

I'm honestly so proud of how good this AI sample generator is turning out.

ALL red samples are right out of the unreleased model + ALL FX are prompt generated.

(ft. AI choirs from my other model😀)

Teaching AI music theory and melody structure is fun af.

#musicproduction #AI #FLStudio #EDM