Abstract: A system for visually representing user behavior within an interactive voice response (IVR) system of a call processing center generates a complete sequence of events within the IVR system for plural calls to the call processing center, the plurality of calls being recorded from end to end. A call flow of the IVR system is modeled as a non-deterministic finite-state machine, such that a start state of the finite-state machine represents a first prompt of the IVR system, other states of the finite-state machine represent subsequent prompts at which a branching occurs in the call flow of the IVR system, exit conditions are represented as end states, and transitions of the finite-state machine represent transitions between call flow states triggered by data inputted by a user or by internal processing of the IVR system. The complete sequences of events for the plural calls are provided to the finite-state machine to produce a two-way matrix of several counters.

Abstract: A system for assessing an automated call routing system's performance is structured to present a caller calling into the automated call routing system with a call that includes an interactive voice response (IVR) portion and, at an option of the caller, an caller/agent dialog portion. The system operates to monitor calls coming into the automated call routing system and to record end-to-end calls from among the calls coming into the automated call routing system. A sequence of events for the recorded calls is transcribed, including events occurring in the IVR portion and in the caller/agent dialog portion, to identify predetermined call events of interest. The transcribed calls are analyzed to infer a complete event sequence of each of the recorded calls. From the analysis, parameters are calculated corresponding to cost effectiveness and usability of the automated call routing system, based at least in part upon information obtained by the caller during the IVR portion.

Abstract: A system for forming a database indexed with content-based parameters, corresponding to recordings of contacts to a contact center, includes an analysis unit, a mining unit, and a correlation unit. The analysis unit performs automated-portion analysis techniques to ascertain predetermined events occurring in recordings of automated portions of the contacts to the contact center. The mining unit performs audio mining of recordings of agent portions of the contacts to the contact center. The correlation unit correlates an event occurring in an agent portion of the recordings with a corresponding one of a plurality of content-based parameters selected to be used as indices of the database, and makes an entry in the database based on the correlated event and the corresponding content-based parameter.

Abstract: An automation benefit of an automated call processing center is quantified as saved agent-time by gathering event-sequence data for a plurality of calls received by the call processing center; calculating an amount of time a task would take if performed by an agent instead of by an automated unit, wherein the task is capable of being performed by the automated unit; examining the event-sequence data to determine whether the task, which is capable of being performed in the automated unit, actually was performed in the automated unit for the plurality of calls; determining an automation rate for the task by calculating what percentage of calls involving the task actually was performed by the automated unit; and calculating an automation benefit by multiplying the automation rate by an amount of agent-time saved by performing the task in the automated unit instead of by an agent, wherein the amount of agent-time saved corresponds to the amount of time the task would take if performed by an agent.

Abstract: A system for analyzing the performance of an interactive voice response (IVR) system operates to record from end to end plural calls from callers. For a plurality of calls, the system codes and annotates an agent-caller interaction, including a topic discussed during the interaction; detects a routing operation of the IVR system by detecting a sequence of events in the IVR system, with reference to a call-flow file that summarizes the call-flow model of the IVR system, for each call; and cross-tabulates routing information obtained from the detected sequence of events with results from the coded and annotated agent-caller interactions.

Abstract: A system for analyzing an interactive voice response (IVR) system of a call processing center determines a complete sequence of events occurring within the IVR system. The IVR system being operable to automatically accept calls from callers and respond to input from the callers. The system models a call flow of the IVR system as a non-deterministic finite-state automaton. A recording of a plurality of calls to the call processing center is split, such that each of plural calls to the call center is in one audio file. For a plurality of calls, the system detects a complete sequence of DTMF and/or speech input to the IVR system, detects selected prompts issued by the IVR system, and inputs the detected sequence of DTMF and/or speech input as well as the detected prompts issued by the IVR system into the finite-state automaton to determine a call-event sequence for that call. The call-event sequence includes information regarding how that call left the IVR system.

Abstract: The method of the present invention is comprised of the steps of locating an error within a recognition hypothesis. A repair hypothesis is generated for the located error from a secondary input signal. At least a portion of the recognition hypothesis is correlated with the repair hypothesis to produce a new hypothesis for the located error. The located error is replaced with the new hypothesis. It has been determined that when the primary and secondary input signals are cross-modal, the chances that the error will be repeated are so low that the correlation step becomes optional.