The Advanced SPICE Model for electrostatic discharge diodes (ASM-ESD) is the industry-standard compact model for ESD diodes. Traditional parameter extraction techniques for ASM-ESD require expert knowledge and can take several minutes to hours to finish. This paper presents a fast and accurate deep learning (DL) based methodology for extracting the ASM-ESD Forward I-V parameters for the first time. The DL parameter extractor is trained using Monte Carlo simulation of transmission-line pulse (TLP) I-V transient simulation. Over 80 thousand data points are generated to train the DL parameter extractor. Different DL structures are evaluated on unseen ASM-ESD simulated data and measured ESD TLP data. The results show that a fully trained DL-parameter extractor can instantly extract parameters within a second, and create a model for the I-V characteristics with excellent accuracy while simplifying the parameter extraction process. This shows the great potential of using DL for developing fast and accurate charge device models for circuit simulation to prevent machine-related failures due to electrostatic discharge.