Korean traditional singing dataset released with detailed vocal pitch labels
Gugak-VocalSet: A Multi-Genre Studio-Quality Dataset of Korean Traditional Singing with Sigimsae Annotations
SoundArtificial IntelligenceDatabasesDigital Libraries
Summary
Capturing the unique sounds of Korean traditional singing is hard because it involves special pitch movements called sigimsae. The authors created Gugak-VocalSet, a large collection of studio recordings from professional singers with detailed labels describing these pitch expressions. This dataset helps computers learn to recognize and analyze Korean singing styles more accurately. They also tested different methods for identifying these pitch details and found some that work well.
What this means in practice
- •For music technologists: Build tools that recognize and analyze Korean traditional vocal techniques using detailed sigimsae labels from the dataset.
- •For sound engineers: Develop high-quality audio processing algorithms tailored to Korean traditional singing’s unique pitch patterns.
Authors
Danbinaerin Han, Wonil Kim, Seah Hong, Dongjoo Moon, Jaeok Lee, Jongpil Lee, Chaewon Kim, Juhan Nam
Abstract
We present Gugak-VocalSet, the first publicly available audio dataset for Korean traditional singing with dense annotations of sigimsae, a structured pitch contour system of vocal expressions. Gugak-VocalSet contains 24.5 hours of unaccompanied solo vocal recordings across 305 tracks of representative repertoire curated by the National Gugak Center, performed by 20 professional singers in a controlled studio environment with fixed key and tempo. In addition to genre, lyrics, and mood/timbre tags, the dataset provides a unified sigimsae ontology and approximately 68,000 segments (71,700 instances including multi-label) of audio-grounded sigimsae annotations; the ontology adopts a functional-position-based hierarchy designed for consistent application across genres, and 14 expert annotators specializing in Korean traditional music performed segment-level labeling accordingly. Baseline experiments on sigimsae classification and event detection show that domain-informed F0 representations remain competitive with, and on classification surpass, frozen probing of a general-purpose music foundation model, while pitch and spectral cues play complementary roles in event-level detection. Companion page: https://www.danbinaerin.com/Gugak-VocalSet/