State-Time-Alignment Phone Clustering Based Language-independent Phone Recognizer Front-end for Phonotactic Language Recognition

Abstract

The now-acknowledged sensitive of Phonotactic Language Recognition (PLR) technology to the performance of the phone recognizer front-end have spawned interests to develop many methods to improve it. In this paper a state-of-art State-Time-Alignment (STA) phone clustering approach to build language-independent phone recognizer is proposed in phonotactic language recognition system to balance the performance and the complexity of the speech tokenizing processing in PLR.Experiments are carried out on the database of National Institute of Standards and Technology language recognition evaluation 2009 (NIST LRE 2009) and the experimental results have confirmed that phonotactic language recognition system using the collaborated language model yields 1.84%, 5.55% and 16.82% in equal error rate (EER), which show that the STA phone clustering based phone recognizer front-end outperforms the original English and Mandaren phone recognizers and other phone clustering methods based phone recognizer.