Viotech detects violin techniques precisely without note segmentation
VioTech: Frame-Level Violin Playing Technique Detection with Multi-Scale Mixture-of-Experts
SoundArtificial IntelligenceMachine LearningMultimedia
Summary
It is hard to tell which playing techniques a violinist uses because violin sounds blend together and notes don't have clear beginnings and ends. The authors created VioTech, a new system that looks at tiny moments of sound (every 10 milliseconds) instead of waiting for full notes. This helps detect different violin techniques more accurately and works well even on new recordings. They also made a new dataset with real violin performances to train and test their system.
What this means in practice
- •For music transcription developers: Improve automatic violin music transcription by detecting playing techniques without needing exact note boundaries.
- •For digital music production teams: Add detailed violin technique recognition to music software for enhanced editing and effects control.
Authors
Ting-Kang Wang, Vincent K. M. Cheung, Nan-Tien Lai, Li Su
Abstract
Playing technique is crucial for expressive violin performance, yet few violin technique detection models exist. Current systems mainly rely on note-level pipelines that first estimate note boundaries and then classify techniques. This is especially problematic for bowed string instruments such as the violin, where soft onsets and continuous bow-string interaction render note segmentation unreliable. Here, we propose VioTech, a frame-level violin technique detection framework that predicts techniques directly at 10 ms resolution without explicit note segmentation. To handle the large temporal-scale variation among techniques, VioTech combines a frozen MERT encoder with a multi-scale Mixture-of-Experts (MoE) architecture and task-specific routing for tonal, articulation, and legato-related predictions. We also introduce VioTech120, the first manually annotated dataset of violin playing techniques constructed from real solo performance recordings, as well as an accompanying note-labeling user interface. Results demonstrate that VioTech achieved consistent improvements over prior methods and significantly enhanced cross-dataset generalization, establishing a new state-of-the-art in violin technique-aware transcription.