Hearing Aid Audio Separation for Streamed and Environmental Voices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users wearing hearing aids face difficulties in distinguishing between streamed content voices and environmental sounds due to overlapping and delayed audio signals, leading to hearing challenges in reverberant environments.

Innovation Solution

An acoustic processing device and system that utilizes feature information to selectively process environmental sounds based on specific voice characteristics, enabling clear hearing of both streamed content and ambient voices by integrating sound collection, reception, processing, and output units.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If external sound is blocked to avoid overlap with streamed content, then auditory clarity of streamed content is improved, but the user cannot hear necessary environmental sounds such as spoken voices

Engineering Contradiction:
Improveauditory clarityVSAvoidhearing capability
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent extracts and removes the specific voice component from environmental sounds using voice extraction processing based on feature information. This allows the hearing aid to eliminate only the interfering spoken voices while preserving other necessary environmental sounds, resolving the contradiction between blocking external sounds and maintaining hearing capability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different processing to different sound components: streamed content receives noise reduction processing while environmental sounds undergo voice extraction processing. This localized differential processing allows clear hearing of both streamed content and necessary environmental sounds simultaneously.

Inventive Principle:
Principle #3Local quality

2Reliability

If streamed content is distributed to the hearing aid, then the user can hear content voice, but overlap with captured environmental voice makes it difficult to hear clearly

Engineering Contradiction:
Improvecontent deliveryVSAvoidvoice distinction
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies different processing methods to different sound sources: noise reduction processing for streamed content and voice extraction processing for environmental sounds. This differential local processing enables clear distinction between content voice and environmental voice even when both are present.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent introduces feature information as an intermediary element that enables the hearing aid to identify and separate specific voices in environmental sounds. This mediator allows the system to distinguish between content voice and environmental voice for appropriate processing.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If streaming is used to deliver content to the hearing aid, then content accessibility is improved, but delay and temporal deviation cause overlapping sounds at different times

Engineering Contradiction:
Improvecontent accessibilityVSAvoidaudio synchronization
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent uses feature information as a mediator that enables real-time identification and processing of voices. This allows the system to handle temporal deviations caused by streaming through real-time voice extraction and synchronization, maintaining audio alignment despite transmission delays.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4694207A1Acoustic processing device, information transmission device, and acoustic processing system
Publication Date: 2026.02.11 SONY GROUP CORP
  • EP4694207A1 patent drawingFigure 1
  • EP4694207A1 patent drawingFigure 2
  • EP4694207A1 patent drawingFigure 3

AI summary

Provided is an acoustic processing device worn on a body of a user, the acoustic processing device including: a sound collection unit that acquires an environmental sound around the user; a reception unit that receives feature information for a specific voice included in a voice output from an acoustic output device to the user; a processing unit that performs acoustic processing on the environmental sound around the user collected by the sound collection unit based on the feature information; and an output unit that outputs the environmental sound processed by the processing unit to the user.