Applied Artificial Intelligence (Dec 2024)
Enhancing Vocal Performance using Variational Onsager Neural Network and Optimized with Golden Search Optimization Algorithm
Abstract
ABSTRACTThe Singing Muscle Ability Training System (SMATS) represents a cutting-edge technological solution dedicated to honing an individual’s vocal prowess. By employing a sophisticated combination of techniques, exercises, and methodologies, this system strategically targets and cultivates the muscle groups integral to singing. In its innovative approach, SMATS integrates Artificial Intelligence (AI) to elevate vocal capabilities further. The proposed SMATS-AI-VONN-GSOA stands out by incorporating the Variational Onsager Neural Network (VONN) alongside the Golden Search Optimization Algorithm (GSOA). This amalgamation tailors training routines based on the user’s progress and preferences, emphasizing the development of muscle memory and control for enhanced vocal performance. Noteworthy is the system’s capacity to analyze data through AI, enabling the creation of personalized training plans. In comparative evaluations, the SMATS-AI-VONN-GSOA method demonstrates a significant performance boost, surpassing existing methods like SMATS-AI-CNN, SMATS-AI-DNN, and SMATS-AI-DNN-HSV by 26.78%, 29.55%, and 21.41% in accuracy, respectively.