Embodiments of the invention provide for the automated retrieval of contextual relevant data in an in-person meeting. In an embodiment of the invention, a method for the automated retrieval of contextual relevant data in an in-person meeting includes capturing speech audio in an in-person meeting through a transducer coupled to a personal computing device and speech recognizing the captured speech audio without regard to any speaker of the captured speech audio so as to produce a corpus of speech recognized text. The method also includes parsing the corpus of speech recognized text in order to identify a multiplicity of keywords and mapping the keywords to one or more records of an enterprise application. Finally, the method includes displaying a user interface to the enterprise application in the personal computing device and displaying in the user interface the one or more records.