In embodiments, a method includes detecting, by a computing device, open applications of the computing device; storing, by the computing device, a buffer that tags and tracks audio content and audio context of the open applications; receiving, by the computing device, a user request to take an action regarding at least one of the open applications; determining, by the computing device, a match between the user request and the at least one of the open applications utilizing the buffer; and initiating, by the computing device, a function based on the user request in response to determining the match.