Papers
1
Total Citations
3
H-Index
1
About
Juhan Nam is a leading researcher in music information retrieval and audio signal processing, with a particular focus on singing voice analysis and deep learning applications. He is best known for creating the Children's Song Dataset (CSD), an open-source resource that provides 200 audio recordings of 50 Korean and 50 English children's songs, sung by a professional pop singer in two different keys. This dataset has become a valuable benchmark for singing voice research, enabling advances in pitch tracking, vocal separation, and music transcription. With over 3 citations on this work alone, Nam's contributions have helped bridge the gap between traditional musicology and modern machine learning. His research also extends to automatic music tagging, instrument recognition, and audio source separation, where he has published influential papers in top venues like ISMIR and ICASSP. By combining rigorous dataset creation with innovative algorithmic approaches, Nam has established himself as a key figure in the field, inspiring both students and fellow researchers to explore the intersection of music and artificial intelligence.
Research Focus
Key Achievements
Top Papers
- 1CSD: Children's Song Dataset for Singing Voice Research3 citations · 2021