Apparatus and method for event-driven content analysis
First Claim
Patent Images
1. An apparatus for event-driven content analysis of an audio interaction captured in a call center, within a computerized system having a processing unit and a storage unit, the apparatus comprising the elements of:
- an audio or video recording device for recording the audio interaction and obtaining an interaction media;
a pivot spot defining component for automatically marking an at least one time position in the audio interaction that indicates the occurrence of an at least one pre-defined event or data item;
a first audio content analysis component of a first audio content analysis type;
a region of interest defining component for defining an initial region of interest, by determining the time limits of an at least one segment of the audio interaction, the segment containing the time position of a pivot spot, and for activating the first audio content analysis component on the initial region of interest for dynamically reducing the time limits of the initial region of interest to obtain the region of interest; and
a second audio content analysis component of a second audio content analysis type different from the first audio content analysis type, for analyzing the region of interest of the audio interaction,wherein the first audio content analysis type and the second audio content analysis type are selected such that the second audio content analysis type requires more computing resources than the first audio content analysis type.
3 Assignments
0 Petitions
Accused Products
Abstract
An apparatus and method for the event-driven analysis of media contents derived from customer interactions is disclosed. Content analysis is executed exclusively on those segments of the interaction media that are relevant in a given context. The steps of the analysis are carried out either in a static or dynamic manner where less demanding on resources analysis type is performed prior to more demanding analysis type.
119 Citations
31 Claims
-
1. An apparatus for event-driven content analysis of an audio interaction captured in a call center, within a computerized system having a processing unit and a storage unit, the apparatus comprising the elements of:
-
an audio or video recording device for recording the audio interaction and obtaining an interaction media; a pivot spot defining component for automatically marking an at least one time position in the audio interaction that indicates the occurrence of an at least one pre-defined event or data item; a first audio content analysis component of a first audio content analysis type; a region of interest defining component for defining an initial region of interest, by determining the time limits of an at least one segment of the audio interaction, the segment containing the time position of a pivot spot, and for activating the first audio content analysis component on the initial region of interest for dynamically reducing the time limits of the initial region of interest to obtain the region of interest; and a second audio content analysis component of a second audio content analysis type different from the first audio content analysis type, for analyzing the region of interest of the audio interaction, wherein the first audio content analysis type and the second audio content analysis type are selected such that the second audio content analysis type requires more computing resources than the first audio content analysis type. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)
-
-
14. A method for event-driven content analysis, within a computerized system having a processing unit and a storage unit, the method comprising the steps of:
-
receiving an audio interaction media between an organization and a customer, the interaction media associated with an at least one event, the interaction media recorded by an audio or video recording device; determining an at least one pivot spot, being a time position, on the interaction media; determining the time limits of the at least one segment of the interaction media to be analyzed, said limits defining an initial region of interest within the interaction; reducing the initial region of interest by performing an at least one first audio content analysis of a first audio content analysis type on the initial region of interest and reducing the initial region of interest in accordance with a result of the at least one first audio content analysis, to obtain a region of interest; and performing an at least one second audio content analysis of a second audio content analysis type on the region of interest, wherein the second audio content analysis type is different from the first audio content analysis type and wherein the first audio content analysis type and the second audio content analysis type are selected such that the second audio content analysis type requires more computing resources than the first audio content analysis type. - View Dependent Claims (15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-
Specification