Chris Donahue
@chrisdonahuey
Gen AI for **human** creativity in music + more. Assistant prof at CMU CSD, 🎼 G-CLef lab. Part time Google DeepMind, Magenta (views my own)
Excited to share SingSong, a system which can generate instrumental accompaniments to pair with input vocals! 📄arxiv.org/abs/2301.12662 🔊g.co/magenta/singso… Work co-led by myself, @antoine_caillon, and @ada_rob as part of @GoogleMagenta and the broader MusicLM project 🧵
Heading to Vancouver for ICML✈️🇨🇦Let’s chat about coding agents, evals, and human-AI collab. I’ll also be on the job market this upcoming cycle, looking for TT faculty roles + post-docs. Here's where you'll be able to find me this week👇
Tokenization has been the final barrier to truly end-to-end language models. We developed the H-Net: a hierarchical network that replaces tokenization with a dynamic chunking process directly inside the model, automatically discovering and operating over meaningful units of data
New VST/AU Plugin! 🚨 Play with Lyria RealTime directly from inside your favorite DAW with “The Infinite Crate” 🎧🎶 Like other Lyria RT demos, you can mix together text prompts and other controls to steer the model in real-time. But now with a VST plugin you can feed audio…
I converted one of my favorite talks I've given over the past year into a blog post. "On the Tradeoffs of SSMs and Transformers" (or: tokens are bullshit) In a few days, we'll release what I believe is the next major advance for architectures.
🔥Happy to announce that the AI for Music Workshop is coming to #NeurIPS2025! We have an amazing lineup of speakers! We call for papers & demos (due on August 22)! See you in San Diego!🏖️ @chrisdonahuey @Ilaria__Manco @zawazaw @huangcza @McAuleyLabUCSD @zacknovack @NeurIPSConf
Try *style* prompting Magenta RT with your favorite audio in our Colab demo! colab.sandbox.google.com/github/magenta… *Live input* audio prompting coming soon 👀
Show don't tell... Magenta RealTime now supports audio prompting in addition to text prompting, so you can now use clips of audio as latent anchors to steer generation. youtu.be/vHIf2UKXmp4?si…
✅ End to end generation of expressive performance *audio* from score *images*! An important step towards seamless interaction with computer music systems and a fun collaboration between Dasaem’s group at Sogang University and my group at CMU
🎶Now a neural network can read scanned score image and generate performance audio in end-to-end😎 I'm super excited to introduce our work on Unified Cross-modal translation between Score Image, Symbolic Music, and Audio. Why does it matter and how to make it? Check the thread🧵
📢 Introducing VERSA: our new open-source toolkit for speech & audio evaluation! - 80+ metrics in one unified interface - Flexible input support - Distributed evaluation with Slurm - ESPnet compatible Check out the details wavlab.org/activities/202… github.com/wavlab-speech/…
At #CHI2025 in Yokohama this week 🌸. My first CHI, excited to finally attend. Check out my incoming PhD student Yewon Kim talk about our recent work on multimodal AI for songwriters, and happy to chat with anyone about human AI interaction for music, programming, and creativity!
Come say hi! I'll be presenting Amuse on 🗓️ April 29 (Tue) at 4:20 PM in Annex Hall F205. This whole journey was made possible thanks to the incredible support of @wewantsj and @chrisdonahuey 🌟
@CopilotArena was featured in @SCSatCMU news! Featuring quotes from me, @iamwaynechi, @atalwalkar and @chrisdonahuey 🥳 📖Check out the article here: cs.cmu.edu/news/2025/copi…
What do developers 𝘳𝘦𝘢𝘭𝘭𝘺 think of AI coding assistants? In October, we launched @CopilotArena to collect user preferences on real dev workflows. After months of live service, we’re here to share our findings in our recent preprint. Here's what we have learned /🧵
Out now in Nature Communications! We investigated the neural basis of dynamic categorization in human visual cortex. See thread for more: nature.com/articles/s4146…
Excited to share our new preprint on the neural mechanisms of flexible shape categorization! With @serences & @nuttidarungrat Link: doi.org/10.1101/2023.0… See 🧵👇
What role can AI play in unlocking musical creativity? At #CHI2025 (🥇 Best Paper Award 🥇), we present Amuse: a songwriting tool for musicians that turns photos, sounds, and stories into chords 🎶 📄 arxiv.org/abs/2412.18940 🎧 yewon-kim.com/amuse/
blog.ml.cmu.edu/2025/04/09/cop… How do real-world developer preferences compare to existing evaluations? A CMU and UC Berkeley team led by @iamwaynechi and @valeriechen_ created @CopilotArena to collect user preferences on in-the-wild workflows. This blogpost overviews the design and…