In this era of rapidly growing technological advancements, Artificial Intelligence (AI) integration into the home environment enhances energy efficiency and human comfort. This research focuses on using an AI model to control fan speed through the PWM technique based on the indoor temperature level. Different machine learning algorithms, including Linear SVM (Support Vector Machines), Fine Gaussian SVM, Quadratic SVM, Linear Regression, Medium Tree, Fine Tree, and Robust Linear, are trained using the collected dataset from the home environment. The trained models are then evaluated, and the decision tree models outperformed the other models as they have minimum Root Mean Squared Error (RMSE) for this scenario. This model is then used for real-time prediction to control fan speed. This study contributes to the realization of sustainable and intelligent home environments by offering real-time adjustments to optimize energy consumption and indoor comfort.