Sport-related sudden cardiac death (SRSCD), defined as "death occurring during sport or within one hour of cessation of training", is the leading cause of death in athletes. SRSCD occurs in the presence of underlying cardiovascular diseases, some of which may be identified by processing electrocardiographic recordings acquired during training (TECGs). A fast and accurate processing of TECGs during or immediately after training is challenging since TECGs are typically highly corrupted by noise and interferences, which may jeopardize their interpretation and identification of abnormal morphologies. The present study evaluated the ability of GPU-based Segmented-Beat Modulation Method (GPU-SBMM) to provide a noise-free estimation of TECGs, and to improve the algorithm by GPU acceleration to make it compatible with modern hardware. In this research, 19 6to-10 min TECGs (sampling frequency: 256 Hz), acquired from 8 subjects while performing 4 different exercise tasks (walk, run, low-resistance bike and high-resistance bike), were analyzed. Results indicate that GPU-SBMM application yielded a significant increase of SNR(dB) (from 1±5 dB to 19±5 dB; p<10-12), also when stratifying by exercise tasks. Additionally, a considerable average speedup of 7.67x is achieved using NVIDIA GeForce 740M GPU processor.