Specific energy, defined as the energy consumed by per unit volumetric removal rate, is one of the key metrics used for evaluating the environmental performance of a machining process. Grinding is a traditional machining process, characterized by high specific energy and poor environmental performance. Growing global concerns about climate change and depletion of natural resources have made improvements in the environmental performance of grinding and other machining processes imperative. The first step that should be taken to minimize specific energy in grinding is to develop a model that will enable researchers to estimate and characterize specific grinding energy. Thus, business and engineering decisions can be made to reduce specific grinding energy. This research presents an approach to estimate specific grinding energy by fusing information from multiple low-cost sensors using machine learning techniques. Six sensors are used to monitor total machine current and voltage, AC motor current and voltage, vibration, and acoustic emissions from the process. Time and frequency domain analysis is conducted to extract features from raw sensorial data that are correlated with the specific grinding energy. An artificial neural network (ANN) model and a support vector machine-regression (SVM-R) model were established to fuse the extracted features and estimate specific grinding energy. The approach was experimentally validated using a commercial surface grinding machine. The results of this experimental validation indicate that the ANN model and SVM-R model were able to estimate specific grinding energy using the low-cost sensors with a mean accuracy of 84.96% and 90.36%, respectively.