Two Approaches to Class-Based Language Models for ASR
Raquel Justo, María Inés Torres · Machine learning for signal processing ... · 2007
In this work, we propose and formulate two different approaches for the language model employed in an Automatic Speech Recognition application. Both approaches make use of class-based language models, but taking into account that the classes are made up of segments or sequences of words. Experiments, carried out over a spontaneous dialogue corpus in Spanish, demonstrate the ability of the proposed models to learn the way in which the language is generated.