Multi-speaker speech signal to text signal validation

Abstract

Systems and methods are disclosed comprising instructions to access audio data of a captured interaction involving one or more participating users of a digital conference tool, receive a first narrative summary of the captured interaction comprising a set of first component narratives, generate a second narrative summary comprising a set of second component narratives, identify at least one component narrative among the set of second component narratives corresponding to the participating users, display the second component narratives of the second narrative summary at a user interface of the participating users, receive user feedback data comprising an adjustment to the displayed text content of the at least one second component narrative, generate a third narrative summary using the user feedback data received from each participating user of the captured interaction, and display the third narrative summary for the captured interaction at the user interface of each participating user.