Fast Search Algorithms for Connected Phone Recognition Using the Stochastic Segment Model.

摘要

In this paper we present methods for reducing the computation time of joint segmentation and recognition of phones using the Stochastic Segment Model (SSM). Our approach to the problem is twofold: first, we present a fast segment classification method that reduces computation by a factor of 2 to 4, depending on the confidence of choosing the most probable model. Second, we propose a Split and Merge segmentation algorithm as an alternative to the typical Dynamic Programming solution of the segmentation and recognition problem, with computation savings increasing proportionally with model complexity. Even though our current recognizer uses context-independent phone models, the results that we report on the TIMIT database for speaker independent joint segmentation and recognition are comparable to that of systems that use context information.