Audio sensors and low-complexity recognition technologies for soldier systems

The use of audio sensors for soldier systems is examined in detail. We propose two applications, speech recognition and speaker verification. These technologies provide two functions for a soldier system. First, the soldier has hands-free operation of his system via voice command and control. Second, identity can be verified for the soldier to maintain the security integrity of the soldier system. A low complexity, high accuracy technology based upon averaging a discriminative classifier over time is presented and applied to these processes. For speaker verification of the soldier, the classifier is used in a text-prompted mode to authenticate the identity of the soldier. Once the soldier has been authenticated, the interface can then be navigated via voice commands in a speaker independent manner. By using an artificial neural network structure, a high degree of accuracy can be obtained with low complexity. We show the resulting accuracy of the speaker verification technology. We also detail the simulation of the speech recognition under various noise conditions.