Hookpad Aria: A Copilot for Songwriters
Chris Donahue, Shih-Lun Wu, Yewon Kim, Dave Carlton, Ryan Miyakawa, John Thickstun
TL;DR
This paper introduces Hookpad Aria, a Copilot-like AI system integrated into Hookpad to assist in writing Western pop songs through symbolic music generation. It supports non-sequential generation modes—left-to-right, fill-in-the-middle, and both melody-to-harmony directions—conditioned on global attributes like meter, key, and tempo. The system is powered by the Anticipatory Music Transformer, a multi-instrument symbolic music LLM with $360$M parameters, fine-tuned on $50{,}000$ lead sheets from TheoryTab, modeling notes as events $\mathbf{e}$ and controls $\mathbf{c}$ with $P_{\theta}(\mathbf{e} \mid \mathbf{c})$ to learn beat-aware, anticipatory generation up to $5$ seconds. From March 2024, Hookpad Aria has generated $318{,}000$ suggestions for $3{,}000$ unique users, with $74{,}000$ accepted into songs, demonstrating practical adoption and a scalable data flywheel for music co-creation.
Abstract
We present Hookpad Aria, a generative AI system designed to assist musicians in writing Western pop songs. Our system is seamlessly integrated into Hookpad, a web-based editor designed for the composition of lead sheets: symbolic music scores that describe melody and harmony. Hookpad Aria has numerous generation capabilities designed to assist users in non-sequential composition workflows, including: (1) generating left-to-right continuations of existing material, (2) filling in missing spans in the middle of existing material, and (3) generating harmony from melody and vice versa. Hookpad Aria is also a scalable data flywheel for music co-creation -- since its release in March 2024, Aria has generated 318k suggestions for 3k users who have accepted 74k into their songs. More information about Hookpad Aria is available at https://www.hooktheory.com/hookpad/aria