Ai-based self-adaptive rhythm electronic piano accompaniment system

An artificial intelligence based (AI-based) self-adaptive rhythm electronic piano accompaniment system, comprising:

• a control unit:
• configured to receive and process information from all modules, and transmit processed results to relevant modules for further processing;
• a single-key trigger module:
• provided with a trigger key for users to activate accompaniment during performance;
• a sensing sensor unit is provided below the key to detect the force applied when the key is pressed;
• a velocity-control signal conversion module:
• configured to receive and process signals from the sensing sensor and convert them into corresponding music control signals;
• a storage module:
• configured to store accompaniment note sequence data of songs, as well as timbre parameter data, preset resonance parameters, parameters of velocity-control signal mapping formulas, and parameters of a timbre synthesis model;
• an AI-based accompaniment generation module:
• configured to retrieve accompaniment note sequence data of corresponding songs from the storage module based on the triggered single key and its timing sequence, and generate real-time accompaniment or harmonies;
• a pitch calculation unit:
• configured to calculate accompaniment pitch based on note numbers and accidentals in the accompaniment note sequence;
• a timbre parameter adjustment module:
• configured to adjust parameters required for timbre generation based on the accompaniment note sequence output by the AI-based accompaniment generation module, the pitch calculated by the pitch calculation unit, and the velocity signal;
• a timbre generation module:
• configured to calculate and synthesize timbre data according to parameters output by the timbre parameter adjustment module;
• a dynamic pitch range adjustment module:
• capable of dynamically adjusting the pitch range output of the timbre generation module based on the triggered note range;
• an audio output module:
• configured to convert the timbre data into sound and output it.