2026

As of August 01, 2026:

• 9 Ongoing Projects
  • Audiovisual Piano Transcription
  • Intelligent Music Accompaniment
  • Human-aligned Music Generation
  • Reasonable AI Music Evaluation
  • PianoVAM Enrichment & Expansion
  • Next-Generation Musical Experience (Confidential)
  • Live Music Understanding (Confidential)
  • Context-driven Music Captioning
  • Audio-based Pedestrian Detection
• Collaborating Professors
  • Alexander Lerch (Georgia Tech)
  • Chris Donahue (CMU)
  • Juhan Nam (KAIST)
  • Shinji Watanabe (CMU)
  • Subhrajit Guhathakurta (Georgia Tech)
• Collaborating Grad Students
  • Nathan Pruyne (CMU)
  • Chaeyeon Han, Eun Ji Oh, Jun Woo Beck, Sangheon Park (Georgia Tech)
  • Hoyeol Sohn, Joonhyung Bae, Junwon Lee, Carolina Carusi (KAIST)
  • Yinghao Ma (QMUL)
• Collaborating Undergrad Students
  • Sancho Gatungay (Georgia Tech)
  • Junhyung Park (KAIST)
  • Haiwen Xia (Peking University)

2026

• CueSheet: Live Demo of Multi-Genre Concert Segment Detection
• Multi-Task Multi-Frame Visual Piano Transcription
• Improving Text-to-Music Generation with Human Preference Rewards
• TuneJury: An Open Metric for Improving Music Generation Preference Alignment

2025

• Audio‑Based Pedestrian Detection in the Presence of Vehicular Noise
• Music Arena: Live Evaluation for Text-to-Music
• The Secret Source: Supporting the Artists Whose Works Have Made Our Works Possible
• Two Web Toolkits for Multimodal Piano Performance Dataset Acquisition and Fingering Annotation
• PianoVAM: A Multimodal Piano Performance Dataset
• NeoLightning: A Modern Reimagination of Gesture-Based Sound Design

2024

• Towards Robust Transcription: Exploring Noise Injection Strategies for Training Data Augmentation