Momentum is a technique that is widely used to improve convergence rates during gradient descent. In this research, we experiment with adding momentum to the Baum-Welch expectation-maximization algorithm for training Hidden Markov Models (HMM). We compare discrete HMMs trained with and without momentum on English text and malware opcode data. The effectiveness of momentum is determined by measuring the changes in model score and classification accuracy due to momentum, as a function of the Baum-Welch iteration. Our extensive experiments indicate that applying momentum to Baum-Welch can accelerate convergence, in the sense of reducing the number of iterations required for initial convergence, particularly in cases where the model is otherwise slow to converge. However, momentum does not seem to improve the final model performance in cases where a sufficiently large number of iterations are used.