Stream-Processing Call Analytics for Real-Time Audio Insights

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional call analytics systems experience delays in providing insights from call audio data, missing opportunities for businesses to intervene in customer interactions and maintain positive experiences, as they rely on delayed analysis of quality, intent, and sentiment.

Innovation Solution

A call analytics system that converts call audio data into lossless messages for real-time processing using a distributed stream-processing platform, such as Apache Kafka, enabling immediate analysis and intervention during calls, and storing data for supplemental analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If conventional call analytics systems process call audio data, then analysis insights are generated, but there is a delay of minutes between data collection and insight availability

Engineering Contradiction:

Improvedelay between data collection and insight availabilityVSAvoidspeed of providing conversational insights

Core Design Contradiction:

Loss of timeVSProductivity

Solution Approach 1:

The system performs preliminary actions by converting call audio data into lossless message format during the call itself, preparing the data for analysis before the call ends. This allows the stream-processing platform to generate insights during or immediately after the call, rather than waiting minutes after call termination.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces conventional batch processing systems with a stream-processing platform that processes call audio data in real-time as it flows through the system. This substitution of processing mechanics enables continuous analysis during the call rather than delayed batch analysis after call completion.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If call audio data is processed using conventional systems, then analysis is performed, but businesses miss opportunities to intervene in customer interactions

Engineering Contradiction:

Improveability to intervene in customer interactionsVSAvoidtime to act on conversational insights

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system implements feedback by delivering conversational insights back to the business system during or immediately after the call, enabling real-time or near-real-time intervention. The stream-processing platform continuously monitors call data and provides feedback loops that allow businesses to act on insights while the customer interaction is still active or just concluded.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

By preparing and processing call audio data during the call itself rather than after termination, the system enables businesses to take preliminary actions and intervene in customer interactions before the call ends, capturing opportunities that would otherwise be lost.

Inventive Principle:

Principle #10Preliminary action

3Manufacturing precision

If VoIP call audio data is handled using conventional methods, then data is processed, but audio quality deteriorates due to lossy compression

Engineering Contradiction:

Improveaudio qualityVSAvoiddata loss from lossy compression

Core Design Contradiction:

Manufacturing precisionVSLoss of substance

Solution Approach 1:

The system converts the harmful effect of lossy VoIP compression into a benefit by implementing a two-stage process: first capturing the compressed VoIP data as received, then converting it to lossless format for analysis. This approach preserves the original compressed data integrity while eliminating compression artifacts for the analysis stage, turning the limitation of VoIP into an opportunity for enhanced quality processing.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent introduces an intermediary conversion process that transforms lossy VoIP audio data into lossless message format. This intermediary step acts as a mediator between the compressed source data and the analysis requirements, preserving audio quality by converting to a lossless representation suitable for accurate transcription and analysis.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11522993B2Systems and methods for rapid analysis of call audio data using a stream-processing platform

Publication Date: 2022.12.06 MARCHEX

AI summary

A call analytics system and associated methods that can be used to rapidly analyze call data and provide conversational insights. The call analytics system receives audio call data of a phone call between a customer and an agent of a business, and converts the call data into one or more messages for handling by a distributed stream-processing platform. In some embodiments, the stream-processing platform is the Apache Kafka platform. The distributed platform processes the messages and communicates with various software modules to generate a variety of conversational insights. When processed by a stream-processing platform, certain analyses can occur in parallel which allows conversational insights to be provided to the businesses shortly (e.g., within seconds) after the call data is received.