Enhanced Recognition Accuracy with the Simultaneous Use of Three Automated Speech Recognition Systems

Abstract

Two studies were performed to test the efficacy of using three different automated speech recognition devices in parallel to obtain speech recognition accuracies better than those produced by each of the individual systems alone. The first experiment compared the recognition accuracy of each of the three individual systems with the accuracy obtained by combining the data from all three systems using a simple “Majority Rules” algorithm. The second experiment made the same comparison, but used a more sophisticated algorithm developed using the performance data obtained from experiment 1. Results from the first experiment revealed a modest increase in speech recognition accuracy using all three systems in concert along with the Simple Majority Rules (SMR) algorithm. Results from the second experiment showed an even greater improvement in recognition accuracy using the three systems in concert and an Enhanced Majority Rules (EMR) algorithm. The implications of using intelligent software and multiple speech recognition devices to improve speech recognition accuracy are discussed.