State of Charge (SOC) is essential to know the amount of energy left in a battery and to estimate the remaining range of the electric vehicle. Conventional methods for SOC calculation like the look-up table method have the drawback of low accuracy. Model-based techniques for SOC estimation consume a lot of time and need a domain expert. Data Driven techniques have proven to be more efficient in predicting the SOC along with great accuracy. Feed Forward Neural network (FNN) model and Long Short-Term Memory (LSTM) model can directly map the Voltage, Current, and Temperature to SOC removing the internal parameters of a battery in between making them simple. In this paper, FNN and LSTM models are employed to estimate the real-time SOC of the battery and their performance is compared based on Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and maximum error.