High price volatility can directly affect the electricity market stability in smart grids. Thus, effective and accurate price forecasts must be implemented to avoid the serious consequences of price dynamics. This study proposes two intelligent techniques to tackle the Electricity Price Forecasting (EPF) problem using machine learning. Firstly, a Support Vector Regression (SVR) model is used to predict the hourly-price. Secondly, a Deep Learning (DL) model is implemented and compared with the SVR model. The results show that the two proposed models are effective tools for EPF. However, the DL approach outperforms the SVR model, with average root mean square error value of 1.1165 and 0.416 respectively.