Psychological Evaluation of Nervousness in Symphony Performance Based on Artificial Intelligence Technology

For symphony orchestra players, controlling stage-performance nervousness is essential for maintaining technical stability and performance quality. Traditional psychological assessment methods are limited by subjective judgment, static questionnaires, and the inability to capture real-time physiological changes. To improve the timeliness and objectivity of psychological assessment, this study develops an artificial-intelligence-based evaluation method for nervousness in symphony performance. Physiological signals and behavioral data are collected through wearable and wireless sensing devices, and a convolutional neural network combined with recurrent neural network modeling is used to analyze multi-source signals such as EEG, ECG, pulse, facial expression, voice, and gesture. Fourier and wavelet-based signal processing are introduced to extract frequency-domain features related to involuntary tremor and autonomic response. Experimental results show that the proposed model improves assessment accuracy by 7.3% compared with traditional methods. The study provides a practical intelligent assessment framework for performance anxiety and demonstrates the value of electromagnetic-compatible wearable signal acquisition in real-time psychological monitoring.