Improved clustered hierarchical tandem system with bottom-up processing

Abstract

The outputs of multi-layer perceptron (MLP) classifiers have been successfully used in tandem systems as features for HMM-based automatic speech recognition. In a previous paper, we proposed Data-driven Clustered Hierarchical MLP (CHMLP) tandem system yielding improved performance by dividing the complicated global phone classification problem into simpler hierarchical tasks, in which specialized MLPs are trained to classify small clusters of confusing phones in a hierarchical structure. In this paper a bottom-up processing is further proposed to enhance the classification in the above CHMLP and offer even better performance. MLP rescoring for the tandem system is also investigated. The best result achieved 19.1% relative error reduction over the MFCC baseline.

A hierarchical, context-dependent neural network architecture for improved phone recognition

“…Hence it makes sense to construct a hierarchy of learners, especially if we have a priori knowledge about how the information is structured. Many efforts have been made to create hierarchic phone classifiers, a recent example being described in [12].…”