Exploiting multiple ASR outputs for a spoken language understanding task

Abstract

In this paper, we present an approach to Spoken Language Understanding, where the input to the semantic decoding process is a composition of multiple hypotheses provided by the Automatic Speech Recognition module. This way, the semantic constraints can be applied not only to a unique hypothesis, but also to other hypotheses that could represent a better recognition of the utterance. To do this, we have developed an algorithm to combine multiple sentences into a weighted graph of words, which is the input to the semantic decoding process. It has also been necessary to develop a specific algorithm to process these graphs of words according to the statistical models that represent the semantics of the task. This approach has been evaluated in a SLU task in Spanish. Results, considering different configurations of ASR outputs, show the better behavior of the system when a combination of hypotheses is considered. © 2013 Springer International Publishing.