Referential semantic language modeling for data-poor domains

This paper describes a referential semantic language model that achieves accurate recognition in user-defined domains with no available domain-specific training corpora. This model is interesting in that, unlike similar recent systems, it exploits context dynamically, using incremental processing and limited stack memory of an HMMlike time series model to constrain search.

• Type: Conference
• Year: 2008
• Where: ICASSP
• Authors: Stephen Wu, Lane Schwartz, William Schuler
• Keywords: Domain-specific Training Corpora, ICASSP 2008, Semantic Language Model, Signal Processing, Time Series Model