Positioning left-hand movement in violin performance: system and user study on fingering pattern ...
summary
An automated fingering system trains BiLSTM modules for string, hand position, and finger on a new dataset of 10 pieces annotated by 10 professionals yielding over 200k note annotations. It outputs three generation modes basic, lowest, nearest that trade off lowest hand placement, nearest-position continuity, or maximum probability. An 8-subject evaluation using 7-point scales for ease of transition, expression, naturalness and skill shows the lowest mode helps beginners on one excerpt, no significant differences on another, and lowest can beat a baseline.
AI-written summary · about
related talks
11mMachine Learning of Personal Gesture Variation in Music Conducting
16mEnseWing: Creating Instrumental Ensemble Playing Experience for Children with Limited Music Training
41mCreating Intelligent Music Instruments with Machine Learning
34mModelling emotional nuance of musical performance
18mPlaying Fast and Loose with Music Recognition
20mUnimanual Pen+Touch Input Using Variations of Precision Grip Postures
33mBuilding "learn to touch type" glove with ...
16mPentelligence: Combining Pen Tip Motion and Writing Sounds for Handwritten Digit Recognition