This study investigates how men and women differ in their use of acoustic features—pitch, amplitude, and duration—when marking three types of linguistic stress: lexical, phrasal, and contrastive. The experiment involved 36 native English-speaking college students from the mid-Atlantic United States, evenly split between 18 male and 18 female participants. Participants’ speech was analyzed using Praat, with statistical methods including Bayesian ANOVAs, random forest classification, and Bayesian mixed-effects regression employed to assess the significance of these acoustic features and their role in the accuracy of stress marking. The results reveal that while both men and women use all three acoustic features, subtle gender differences exist, particularly in the use of pitch. Women tend to rely on a broader range of acoustic features, especially pitch, for contrastive and lexical stress, whereas men emphasize amplitude and duration. The study also highlights the complexities in human versus machine classification of stress patterns, noting that human coders likely outperform random forest models due to their ability to integrate acoustic information. Overall, these findings contribute to our understanding of how gender influences prosodic elements in speech and complement previous studies by offering a more nuanced analysis using advanced statistical methods.