Abstract: Systems and methods are described herein for improving walking or running performance in a variety of settings, including athletic training and physical therapy settings. Software programs comprising machine learning algorithms are employed with the objective of synthesizing real-time, actionable recommendations for performance improvement based on data inputs from a plurality of sensors incorporated into footwear. The recommendations may be visualized or heard during and after performance, either on a mobile computing device such as a smartphone or smartwatch or over a web interface on a home computer, and historical performance data may be incorporated into a personalized database for each user. The systems include real-time recommendations for changes in form and personalized coaching such as of athletes seeking to improve performance. The systems and methods described herein may also be employed for the purposes of injury prevention or rehabilitation in a medical setting.