KE Mechanism

The KE System follows a five step process: (a) visual layout of the desired menu-tree using a graphical user interface; (b) specification of keywords to serve as synonyms for menu-tree nodes for routing purposes; (c) generation of recognizer and NLU grammars through combination of keywords with a unique garbage model; (d) training of NLU via derivation of semantic relationships between inputted keywords; and (e) testing and tuning of grammars.

The recognizer grammar is loaded inside the 3rd party speech recognition engine and allows detection of specified keywords in a wide variety of natural language utterances. The NLU grammar is loaded inside the Vestec NLU engine and processes the output of the 3rd party recognizer for semantic interpretation and call-routing. Specifically, there is no need to build grammars using conventional tagged transcriptions-based training methodology or to write a semantic interpreter for processing recognition results.