Efficient Segmental Cascades for Speech Recognition

TL;DR

This paper presents methods to improve the efficiency of discriminative segmental models in speech recognition, enabling faster decoding and training without sacrificing performance.

Contribution

It introduces techniques like feature set reduction, frame subsampling, and pruning to make segmental cascades practical and efficient.

Findings

Maintains competitive phonetic recognition performance

Significantly reduces decoding and training time

Effective combination of efficiency techniques

Abstract

Discriminative segmental models offer a way to incorporate flexible feature functions into speech recognition. However, their appeal has been limited by their computational requirements, due to the large number of possible segments to consider. Multi-pass cascades of segmental models introduce features of increasing complexity in different passes, where in each pass a segmental model rescores lattices produced by a previous (simpler) segmental model. In this paper, we explore several ways of making segmental cascades efficient and practical: reducing the feature set in the first pass, frame subsampling, and various pruning approaches. In experiments on phonetic recognition, we find that with a combination of such techniques, it is possible to maintain competitive performance while greatly reducing decoding, pruning, and training time.