Assistance System Speech Intelligibility Acoustic Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing assistance systems are not adequately adapted to users with hearing impairments, leading to reduced effectiveness in noisy environments, as they primarily rely on simple amplification rather than intelligent modulation of speech output to enhance intelligibility.

Innovation Solution

An assistance system that uses sensors to analyze the acoustic environment and the user's hearing capacity, adjusting speech presentation parameters such as frequency, intensity, and modality to optimize intelligibility, including the use of visual cues and gestures, to adapt to both the user's hearing abilities and environmental conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If simple amplification of speech output is used, then the system is easy to implement, but speech intelligibility deteriorates in noisy environments and for hearing-impaired users

Engineering Contradiction:
Improvesystem implementation simplicityVSAvoidspeech intelligibility
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The system dynamically changes multiple parameters of the speech output signal including frequency spectrum distribution, temporal envelope characteristics, and spatial positioning. These parameter modifications are adapted based on the user's hearing profile and the acoustic environment to optimize intelligibility without requiring complex hardware changes

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces simple acoustic amplification with intelligent signal processing that uses computational models of hearing impairment and acoustic environment analysis. This substitution transforms a purely mechanical/amplification-based system into an intelligent adaptive system that processes and modifies speech signals based on multiple input parameters

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If speech output parameters are dynamically adjusted, then speech intelligibility improves for hearing-impaired users, but system complexity increases

Engineering Contradiction:
Improvespeech intelligibilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary analysis of the user's hearing capacity and the acoustic environment before generating speech output. By pre-processing these inputs and determining optimal parameter settings in advance, the system avoids complex real-time adjustments during speech delivery, reducing computational complexity while maintaining intelligibility

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces intermediate processing stages that include acoustic environment analysis modules, hearing profile processing modules, and parameter optimization modules. These intermediaries break down the complex task of adaptive speech generation into manageable sequential steps, each handling a specific aspect of the problem

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of information

If multimodal communication including visual cues is used, then comprehension improves, but device complexity increases

Engineering Contradiction:
Improveinformation comprehensionVSAvoiddevice complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system merges auditory speech output with visual communication modalities such as lip synchronization, facial expressions, and gesture coordination. By combining these modalities into a unified communication output, the system enhances information delivery without requiring separate independent systems for each modality

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3772735B1Assistance system and method for providing information to a user using speech output
Publication Date: 2024.05.15 HONDA RES INST EUROPE
  • EP3772735B1 patent drawingFigure 1
  • EP3772735B1 patent drawingFigure 2~3
  • EP3772735B1 patent drawingFigure 4

AI summary

An assistance system performs a method for providing information to an assisted person using speech output. The system comprises at least one sensor for acoustically sensing an environment, in which the assistance system and the assisted person are located. The system further comprises a processor that is configured to analyze a sensor output from the at least one sensor. The processor estimates a potential interference of an intended speech output with the sensed acoustic environment in the common environment on the basis of the analysis result. Additionally, the processor obtains information on an assisted person's hearing capacity and determines an expected intelligibility of speech output on the basis of the estimated interference and the obtained information on the assisted person's hearing capacity. The system further comprises a speech output signal generation unit. The speech presentation signal generation unit generates a speech presentation signal in accordance with the determined modality for intended speech presentation. The speech output signal is then supplied at least to a loudspeaker or hearing aid for outputting the speech output including the information to be provided to the user.