Automated Speech Processing for Customer Satisfaction Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for analyzing customer satisfaction in telecommunication interactions are inefficient due to low response rates to surveys and non-representative samples, leading to inaccurate assessments and inefficient use of resources.

Innovation Solution

Implementing automated speech processing systems that use machine-learning-based feature extraction and predictive modeling to analyze telecommunication interactions, generating predicted satisfaction classifications based on digitally-encoded speech representations and structured features.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If automated speech processing systems are implemented, then measurement precision of customer satisfaction is improved, but device complexity increases

Engineering Contradiction:

Improvecustomer satisfaction measurement accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual survey analysis with automated machine learning algorithms that process speech data. The system uses trained models to automatically extract features from speech representations and generate satisfaction classifications, eliminating the need for manual survey processing and significantly improving measurement precision while maintaining automated operation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces trained machine learning algorithms as intermediaries between raw speech data and satisfaction measurements. These algorithms act as mediators that automatically process speech representations, extract relevant features, and generate predicted satisfaction classifications, resolving the contradiction by automating the measurement process while improving accuracy.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If survey methods are used to measure customer satisfaction, then implementation simplicity is maintained, but reliability of satisfaction data deteriorates due to low response rates

Engineering Contradiction:

Improvesatisfaction data representativenessVSAvoiddata collection system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent enables the system to automatically process and analyze speech data without requiring customer participation in surveys. The machine learning models self-service by automatically extracting features from available speech recordings and generating satisfaction measurements, eliminating the reliability issue of low survey response rates while maintaining systematic operation.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the survey-based data collection mechanism with automated speech processing. Instead of relying on customers to complete surveys, the system automatically analyzes speech interactions using trained machine learning algorithms, significantly improving data reliability and representativeness while automating the entire measurement process.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If manual survey processing is used, then resource consumption is low, but productivity of satisfaction analysis deteriorates

Engineering Contradiction:

Improvesatisfaction analysis efficiencyVSAvoidcomputational resource consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent employs pre-trained machine learning algorithms that have already learned from extensive training data. These pre-trained models can quickly process new speech data without requiring intensive real-time computation for training, thereby improving productivity while managing resource consumption efficiently through the use of prepared models.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces manual survey processing with automated machine learning-based speech analysis. The system uses trained algorithms to automatically extract features and generate satisfaction measurements, dramatically improving productivity by processing large volumes of data automatically while the computational resources are managed through efficient algorithm design and pre-trained models.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12211494B2System and method for automated processing of speech using machine learning

Publication Date: 2025.01.28 CONDUENT BUSINESS SERVICES LLC

AI summary

Automated systems and methods are provided for processing speech, comprising obtaining a digitally-encoded speech representation corresponding to a telecommunication interaction, wherein the digitally-encoded speech representation includes at least one of a voice recording or a transcript derived from audio of the telecommunication interaction; obtaining a digitally-encoded data set corresponding to at least one structured feature of the telecommunication interaction; obtaining a reference set, wherein the reference set includes a set of binary-classified existing satisfaction classifications; obtaining a trained machine learning algorithm, wherein the machine learning algorithm has been trained using a first plurality of reference telecommunication interactions which include user-provided satisfaction scores; extracting a feature set from the digitally-encoded speech representation; and by the machine learning algorithm and based on at least one structured feature and the feature set, generating a predicted satisfaction classification for the telecommunication interaction.