Network-based connected digit recognition using explicit acoustic-phonetic modeling

This paper describes a system for speaker-independent connected digit recognition in which explicit acoustic-phonetic features and constraints play a significant role. The digit vocabulary is modeled using a finite-state pronunciation network whose branches correspond to meaningful acoustic-phonetic units. Each branch is associated with an acoustic pattern matcher which employs a combination of whole-spectrum and feature-based metrics. The system has been evaluated using 8621 utterances from the Texas Instruments (TI) multi-dialect, connected digits database. The best configurations of the recognizer achieve string recognition accuracies of 96.5% and 97.8% when the length of the input string is unknown and known, respectively, and when different talkers are used for training and testing.