A multi-source knowledge fusion strategy to improve confidence measure in a lattice-based spoken term detection system

This paper mainly concerns the problem of Confidence measure estimation for Spoken Term Detection (STD). The detecting precision is always a main obstacle to make STD system to be applicable in real-world. In this context, a multi-source knowledge fusion strategy was proposed to improve the qualification of Confidence measure of detected spoken term which is mainly estimated by posterior probability before. For lattice based STD system, a collection of optimal predictive information of detected term is extracted, and the hidden-units Conditional Random Fields (hidden-units CRFs) is adopted to combine these information into a normalized conditional probability to stand for an alternative score of detected term. More precisely, the discriminative ability of multi-source knowledge fusion based Confidence measure is proved to be superior to the posterior based Confidence measure first. And then, the new proposed Confidence measure was combined with the posterior to improve detecting precision and decrease False Alarm rate (FA) in a lattice-based STD system for Conversional Telephone Speech (CTS). Experimental results show that the new proposed Confidence measure has strong complimentary effect and improve the detecting precision about 17% over the baseline system in high precision area. ©2014 Binary Information Press