Versatile and Adaptable Natural Language Understanding Engine
Semantic Interfaces has developed SITA to provide a fast and accurate understanding of text and speech in any given application. SITA facilitates contextual text and voice user interaction by maintaining an internal state machine. SITA can issue context-specific responses during dialogs with users of, e.g., chat bots, simulations and serious games.
Research has shown that contextual information plays an important role in understanding speech. SITA simulates human contextual and linguistic intuitions. It reconstructs incoherent text and voice commands into application specific coherent commands.
Field tests show that SITA can correct up to 70% of errors in automatic speech recognition. SITA is able to correctly understand the user’s intention by using semantic knowledge.
SITA is also well suited for chatbot development as it can easily handle textual variations of the expected user intentions – and prompt when required information is missing. Based on a state machine, it can engage in a meaningful dialog with the user.
SITA can be integrated as an in-premise or web service for natural language understanding.