 The proposed hybrid framework based on the transformer architecture, referred to as the Transformer for Hand Gesture Recognition, TriHGR, improves the accuracy of deep learning based hand gesture recognition, HGR, via surface electromyogram, SEMG, signals by capitalizing on recent advances in hybrid models and transformers. The TriHGR architecture consists of two parallel paths followed by a linear layer that acts as a fusion center to integrate the advantage of each module and achieves recognition accuracies of 86.00 and hashtag X0025-93.74 and hashtag X0025 on the commonly used second MENAPRO dataset, which is higher than state-of-the-art performance for several datasets with different numbers of gestures. This article was authored by Sohail Zabihi, Ilahi Rahimian, Amir Asif, and others.