The Hodgkin-Huxley model is a well-known mathematical model for the generation and propagation of action potentials in neurons. However, determining certain parameters of the model involves a complex combination of experiments and data tuning. Parameter estimation of the Hogkin-Huxley model has been an active research area for several decades, and various optimization techniques have been used including Genetic Algorithms. The Genetic Algorithms have been successful in finding parameter values that fit experimental data well, but they can be computationally expensive and may require significant expertise to tune. More recently, Artificial Neural Networks have been applied to parameter estimation of mathematical models, it can often provide faster and more accurate results than traditional and global optimization methods. In this research, we propose an dense neural network approach for parameter estimation in the Hodgkin-Huxley model, which is trained on a large voltage-clamp type simulations. Our findings showcase its superiority over Genetic Algorithms, in terms of computational efficiency and accuracy.