Performance Adjustment of Speech Rate in Automatic Speech Recognition

Rate of speech (ROS) has a greater influence in both spectral features and word pronunciations that affect automatic speech recognition (ASR) systems. To deal with these ROS effects, the research proposes to use parallel, rate-specific, acoustic models: one for fast speech, the other for slow speech. Rate switching is permitted at word boundaries, to allow modeling within sentence speech rate variation, which is common in conversational speech. The Input Signal Processing configures the front end for ROS to identify the fast and slow speech and customize the application to handle both the situation well..