Hearing loss (HL) is a common public health concern which has implications beyond communication, with potential physiological effects on diverse aspects of health. Audiology and biomechanics have traditionally been distinct fields, but recent research has started to bridge the gap. Studies have hinted at potential correlations between hearing loss and changes in gait, opening up new avenues for exploration. In this study, we investigate the intersection of audiology and biomechanics by exploring the relationship between hearing loss and gait. During a walk test, gait data were collected using advanced sensor technologies. Audiometric measurements at key frequencies (500, 1000, 2000 and 4000 Hz) from both ears were also obtained. Two rigorous analysis pipelines were established: traditional machine learning (ML) pipeline, and deep learning (DL) pipeline. Our traditional ML pipeline utilized ANOVA F-test for feature selection and a battery of ML classification algorithms. In addition, we introduced a deep learning pipeline which directly learned from raw gait data, eliminating the need for manual feature engineering. Our analysis unveiled a compelling relationship between gait and the presence of hearing loss. Notably, random forest classifier emerged as the most effective among ML models with an average area under the receiver operating characteristic curve (AUROC) of 82.11%±8.30%. Concurrently, the deep learning model, leveraging raw gait data, consistently demonstrated an exceptional AUROC exceeding 90%. "Footprints of hearing loss" presents an innovative exploration of the relationship between gait data and hearing loss, showcasing the potential of both traditional ML and DL approaches. This study not only contributes to our understanding of the interplay between specific gait features and hearing loss but also highlights the transformative impact of DL in healthcare. It marks a substantial step toward a deeper insight into the physiological impact of hearing loss and its broader healthcare implications.