AUTOMATED EXTRACTION OF SEMANTIC CONTENT AND GENERATION OF A STRUCTURED DOCUMENT FROM SPEECH

摘要

Techniques are disclosed for automatically generating structured documentsbased on speech, including identification of relevant concepts and theirinterpretation. In one embodiment, a structured document generator uses anintegrated process to generate a structured textual document (such as astructured textual medical report) based on a spoken audio stream. The spokenaudio stream may be recognized using a language model which includes aplurality of sub-models arranged in a hierarchical structure. Each of the sub-models may correspond to a concept that is expected to appear in the spokenaudio stream. Different portions of the spoken audio stream may be recognizedusing different sub-models. The resulting structured textual document may havea hierarchical structure that corresponds to the hierarchical structure of thelanguage sub-models that were used to generate the structured textual document.