Delayed Interactive Auto Attendant IVR Call Filtering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Unwanted automated calls from interactive voice response (IVR) systems and unsolicited SMS text messages are intrusive and difficult for recipients to manage effectively.

Innovation Solution

A Delayed Interactive Auto Attendant (DIAA) system that intercepts incoming calls, analyzes caller ID, and uses speech-to-text, machine learning, and AI to determine if a call is from an IVR system, filters out fraudulent or unwanted calls, and provides responses on behalf of the subscriber if necessary, allowing wanted automated calls to reach the subscriber while managing interactions with IVR systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If all incoming calls are allowed to reach subscribers, then no legitimate calls are missed, but unwanted automated calls cause intrusion and waste

Engineering Contradiction:

Improveintrusiveness of unwanted callsVSAvoiddelivery of legitimate calls

Core Design Contradiction:

Object-affected harmful factorsVSReliability

Solution Approach 1:

The patent introduces an intermediary system (mobile carrier network or third-party service) that sits between the IVR system and the subscriber. This intermediary analyzes incoming calls, identifies IVR-originated calls through various methods (caller ID analysis, speech-to-text conversion, machine learning), and selectively blocks unwanted calls while allowing legitimate ones through. The intermediary acts as a mediator that filters traffic without requiring direct action from the subscriber.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary analysis of incoming calls before they reach the subscriber. By converting speech to text, analyzing caller ID patterns, and using machine learning algorithms in advance, the system determines whether a call is from an IVR system and whether it should be blocked. This preliminary action prevents unwanted calls from ever reaching the subscriber, eliminating intrusiveness before it occurs.

Inventive Principle:

Principle #10Preliminary action

2Object-affected harmful factors

If automated call filtering is implemented, then unwanted calls are reduced, but legitimate automated calls may be incorrectly blocked

Engineering Contradiction:

Improveunwanted automated callsVSAvoidaccuracy of call classification

Core Design Contradiction:

Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms where subscribers can report incorrect classifications (false positives or false negatives). This feedback is used to retrain and improve the machine learning models, making them more accurate over time. The system continuously learns from user interactions to refine its classification of IVR versus human calls, reducing incorrect blocking while maintaining filtration effectiveness.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system uses multiple parameters for call analysis including caller ID patterns, speech-to-text content, tone analysis, and interaction patterns. By changing and weighing different parameters dynamically, the system can adjust its classification accuracy. Machine learning algorithms analyze multiple parameters simultaneously to determine the likelihood of a call being from an IVR system, improving precision in call classification.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If manual review of each call is required, then accurate filtering is achieved, but time and complexity increase significantly

Engineering Contradiction:

Improvecall filtering accuracyVSAvoidcall processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs self-service by automatically analyzing and classifying incoming calls without requiring manual human review. Machine learning algorithms and speech-to-text conversion enable the system to autonomously determine whether a call is from an IVR system and should be blocked. This automation maintains high filtering accuracy while eliminating the time loss and complexity associated with manual call-by-call review.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces mechanical manual review processes with automated electronic systems. Speech-to-text conversion technology transforms audio calls into text for rapid analysis, machine learning algorithms automatically classify calls based on patterns, and computer systems make blocking decisions in seconds. This substitution of mechanical human review with electronic automation maintains precision while dramatically reducing processing time.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11553084B2Delayed interactive auto attendant

Publication Date: 2023.01.10 AT&T INTELLECTUAL PROPERTY I L P

AI summary

Aspects of the subject disclosure may include, for example, a method of receiving, by a processing system including a processor, a call for a subscriber; determining, by the processing system, that the call is from an interactive voice response (IVR) system; determining whether the IVR system is providing a query that requires a response; and responsive to a first determination that the IVR system requires the response, providing the response. Other embodiments are disclosed.