Learning Model Generation for Automatic Percussion Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic musical performance technologies for percussion instruments struggle to produce aurally natural performances solely by moving a stick in accordance with music data, lacking the nuance and variability of human execution.
Innovation Solution
A learning model generation method and device that conduct machine learning on the musical sounds emitted from percussion instruments to generate numerical values for setting musical performance parameters, such as start position, standby period, driving period, and release period, for an automatic musical performance, using a combination of a musical performance control device, detection device, and drive mechanism to adjust and optimize striking parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If music data is used to control stick movement for automatic percussion performance, then automation is achieved, but aural naturalness deteriorates
Solution Approach 1:
The system uses sound detection to capture actual percussion instrument output and feeds this information back to adjust stick movement parameters. The detection device monitors the sound produced, and this feedback is used to dynamically modify the driving signals to actuators, enabling the system to learn and replicate natural playing characteristics while maintaining automation.
Solution Approach 2:
The system performs self-learning by automatically analyzing detected sound patterns and adjusting its own control parameters without external intervention. The learning model generates updated stick movement commands based on the correlation between detected sounds and desired performance characteristics, allowing the automated system to improve its naturalness autonomously over time.
2Manufacturing precision
If machine learning is applied to generate performance parameters from musical sound, then aural naturalness is improved, but device complexity increases
Solution Approach 1:
A learning model acts as an intermediary component that bridges the detection device and the actuator control system. This learning model processes the complex task of correlating detected sounds with appropriate stick movement parameters, isolating the complexity within a dedicated module while keeping the overall system architecture manageable and modular.
3Manufacturing precision
If multiple parameters are adjusted for natural performance, then performance quality is improved, but control complexity increases
Solution Approach 1:
The system dynamically adjusts multiple performance parameters including stick position, velocity, acceleration, and timing based on the learning model's analysis. By systematically varying these parameters and correlating them with detected sound characteristics, the system achieves high performance quality while managing control complexity through structured parameter optimization.
Data Source
AI summary
Disclosed is a learning model generation method executed by a computer, including: striking a percussion instrument with a striking member to emit a musical sound; and conducting machine learning upon receiving an input of the musical sound emitted from the percussion instrument, and generating, based on the machine learning, a learning model for outputting numerical values for setting musical performance parameters for an automatic musical performance of the percussion instrument that is struck when the striking member is driven.