Voice Coaching System for Scalable Speech Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voice coaching systems fail to effectively improve speech and voice competences of users, leading to monotonous and repetitive interactions, lower customer satisfaction, and reduced efficiency in conversations, particularly in call centers.

Innovation Solution

A voice coaching system comprising a device with an interface, processor, and memory, which analyzes audio data to determine speaker metrics, identifies training criteria, and provides personalized coaching sessions to enhance speech skills and customer interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated voice coaching systems are implemented, then training scalability and coverage are improved, but training quality and personalization deteriorate

Engineering Contradiction:
Improvetraining scalabilityVSAvoidtraining quality
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system continuously monitors speaker metrics during conversations and provides real-time feedback through coaching triggers. The feedback loop compares actual speaker performance against ideal speaker behavior models, enabling automated personalization of training interventions based on individual performance gaps while maintaining scalability across large user bases.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts training parameters such as coaching trigger thresholds, feedback frequency, and intervention intensity based on individual speaker performance data. This allows the automated system to adapt training quality to each user's specific needs while maintaining overall system scalability.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If continuous monitoring of speaker metrics is performed, then speech competence improvement is enhanced, but system complexity and computational resources increase

Engineering Contradiction:
Improvespeech competence improvementVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system extracts only the most critical speaker metrics (tone, volume, speech rate, pauses) from the audio data for continuous monitoring, rather than analyzing all possible acoustic features. This selective extraction approach maintains reliable speech competence improvement while reducing computational complexity and resource requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The monitoring system is divided into modular components: audio data acquisition, metric extraction, threshold comparison, and coaching trigger generation. Each module handles specific tasks independently, reducing overall system complexity while enabling continuous reliable monitoring of speaker performance.

Inventive Principle:
Principle #1Segmentation

3Productivity

If automated coaching triggers are generated based on speaker metrics, then training efficiency is improved, but accuracy in identifying training needs deteriorates

Engineering Contradiction:
Improvetraining efficiencyVSAvoidaccuracy in identifying training needs
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system pre-defines multiple coaching trigger conditions based on different speaker metric thresholds and performance patterns before deployment. These pre-configured triggers are refined through iterative testing and validation against ground truth data, enabling the automated system to accurately identify training needs while maintaining high training efficiency.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4020467B1Voice coaching system and related methods
Publication Date: 2025.10.22 GN HEARING AS
  • EP4020467B1 patent drawingFigure 1
  • EP4020467B1 patent drawingFigure 2A
  • EP4020467B1 patent drawingFigure 2B

AI summary

Voice coaching system, voice coaching device, and related methods, in particular a method of operating a voice coaching system comprising a voice coaching device is disclosed, the method comprising obtaining audio data representative of one or more voices, the audio data including first audio data of a first voice; obtaining first voice data based on the first audio data; determining whether the first voice data satisfies a first training criterion; in accordance with determining that the first voice data satisfies the first training criterion, determining a first training session; outputting, via the interface of the voice coaching device, first training information indicative of the first training session.