Our contribution to this year's MATES conference "Eavesdropping Opponent Agent Communication Using Deep Learning" (by T. Gabel, A. Tharwat and E. Godehardt) has been accepted for oral presentation. In this paper, we present a method for learning to interpret and understand foreign agent communication. Our approach is based on casting the contents of intercepted opponent agent communication to a bit-level representation and on training and employing deep convolutional neural networks for decoding the meaning of received messages. We empirically evaluate our method on real-world data acquired from the multi-agent domain of robotic soccer simulation, demonstrating the effectiveness and robustness of the learned decoding models.