For the final project of my Digital Signal Processing class, ECE 564, I developed a complete isolated word recognition system which uses a Mel Frequency Cepstral Coefficient (MFCC) and Dynamic Time Warping (DTW) approach in MATLAB. The system was trained on 10 - 15 examples per word and had a vocabulary of 10 words. While being far computationally cheaper than deep learning or neural network algorithms, the system achieved a 100% accuracy on all test words, making it well-suited for embedded systems with limited computational power.