{"id":193578,"date":"2025-05-08T13:52:39","date_gmt":"2025-05-08T04:52:39","guid":{"rendered":"http:\/\/ee.presscat.kr\/?post_type=research-achieve&p=193578"},"modified":"2026-04-13T07:08:31","modified_gmt":"2026-04-12T22:08:31","slug":"ee-prof-sung-ju-lee-unveils-ai-songwriting-companion-amuse-to-support-music-creation","status":"publish","type":"research-achieve","link":"http:\/\/ee.presscat.kr\/en\/research-achieve\/ee-prof-sung-ju-lee-unveils-ai-songwriting-companion-amuse-to-support-music-creation\/","title":{"rendered":"EE Prof. Sung-Ju Lee Unveils AI Songwriting Companion \u2018Amuse\u2019 to Support Music Creation"},"content":{"rendered":"
\"\"
< (From left) Professor Chris Donahue of Carnegie Mellon University, Ph.D. candidate Yewon Kim of the School of Electrical Engineering at KAIST, and Professor Sung-Ju Lee of the School of Electrical Engineering at KAIST ><\/span><\/figcaption><\/figure>\n

 <\/p>\n

Imagine if music creators had a collaborator who could help brainstorm initial ideas or assist when they hit a creative block or someone who could genuinely support exploring various musical directions. A research team at KAIST\u2019s School of Electrical Engineering has developed an AI tool designed to be such a co-creative partner in music composition.<\/span><\/p>\n

 <\/p>\n

Professor Sung-Ju Lee\u2019s research team at the School of Electrical Engineering has developed an AI-based music creation support system named Amuse. This research was awarded the Best Paper Award, an honor given to only the top 1% of papers, at CHI (ACM Conference on Human Factors in Computing Systems), the world\u2019s leading conference in human-computer interaction, held in Yokohama, Japan from April 26 to May 1.<\/span><\/p>\n

 <\/p>\n

Amuse is an AI-based system that supports songwriting by transforming diverse forms of inspiration, such as text, images, or audio, into harmonic structures (chord progressions). For example, when a user inputs a phrase like \u201cmemories of a warm summer beach,\u201d an image, or a sound clip, Amuse generates and suggests a chord progression that matches the mood and atmosphere of the inspiration.<\/span><\/p>\n

 <\/p>\n

Unlike conventional generative AI, Amuse respects the user\u2019s creative process and offers a flexible, interactive approach that allows users to modify and integrate the AI\u2019s suggestions, encouraging natural and creative exploration.<\/span><\/p>\n

 <\/p>\n

The core technology of the Amuse system is a hybrid generation method. It first uses a large language model to generate music chords based on the user’s textual input. Then, a second AI model trained on actual music data filters out unnatural or awkward results through a process called rejection sampling. These two processes are seamlessly integrated to produce musically coherent outcomes.<\/span><\/p>\n

 <\/p>\n

\"\"
< (Figure 1) System structure of Amuse. Music-related keywords are extracted from user input, and a chord progression is generated using a large language model and refined via rejection sampling (left). Chords can also be extracted from audio inputs (right). The bottom visualizes the harmonic structure of the generated chords. ><\/span><\/figcaption><\/figure>\n

 <\/p>\n

The research team conducted user studies with actual musicians, and the findings suggest that Amuse has strong potential not just as a music-generating tool but as a co-creative partner that enables collaboration between humans and AI.<\/span><\/p>\n

 <\/p>\n

The study, authored by Ph.D. candidate Yewon Kim(KAIST), Professor Sung-Ju Lee(KAIST), and Professor Chris Donahue (Carnegie Mellon University), demonstrates the creative potential of AI systems for both academia and industry.<\/span><\/p>\n