This document presents research on developing a machine learning model to recognize gender from voice recordings. The researchers collected a dataset of over 60,000 audio samples from online repositories. They extracted acoustic features from the recordings like mean, median and standard deviation of dominant frequencies. These features were used to train four machine learning algorithms - Support Vector Machine, Decision Tree, Gradient Boosted Trees and Random Forest. The Random Forest model achieved the highest testing accuracy of 89.3% for gender recognition. Feature importance analysis showed that standard deviation, mean and kurtosis were the most important discriminative features for the tree-based models. The research demonstrates that machine learning can effectively perform voice-based gender recognition using acoustic features extracted from speech.