Lymba Paraphraser
Lymba Paraphraser automatically creates statistical language models (SLM) and context free grammar (CFG). Given a set of categories associated with a directed dialog prompt, Lymba Paraphraser automatically generates statistically probable ways someone might ask for those categories at a prompt, thereby reducing manual labor involved in collecting test user utterances, and decreasing the overall demployment time of a directed dialog system. Further, Lymba Paraphraser is empowered by additional knowledge generated by Jaguar ontologies for customer specific jargon.
Lymba Paraphraser reduces the manual labor eliminates the need for a collection of user utterances, and decreases the development time of generating CFGs or SLMs.
Lymba Paraphaser takes an ASCII file containing a set of categories for a prompt and their descriptions as input and produces a grammar file or statistical n-gram based language model in a format specified in the W3C Speech Recognition Grammar Specification (SRGS).
To learn more about how your organization can benefit from using natural language processing solutions from Lymba, please contact us.