Voice Analysis Device for Mental Disease Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing voice analysis technologies face challenges in accurately estimating mental/nervous system diseases like major depression and bipolar disorder due to environmental acoustic disturbances and the lack of effective biomarkers, leading to inaccurate diagnoses and difficulty in discriminating between these conditions.

Innovation Solution

A device and method that extract acoustic feature amounts independent of the recording environment, using a combination of first and second acoustic feature amounts to calculate prediction values for estimating mental/nervous system diseases, specifically distinguishing between major depression and bipolar disorder through voice analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If voice analysis is performed in a general room environment, then the device can be used conveniently in various locations, but acoustic disturbances from walls, floors, and ceilings alter the acoustic feature amount and degrade estimation accuracy

Engineering Contradiction:

Improvelocation independenceVSAvoiddisease estimation accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent extracts and removes the environmental acoustic characteristics (reflections from walls, floors, ceilings) from the voice signal processing pipeline by focusing only on acoustic feature amounts that do not change with recording environment, thereby isolating the disease-related acoustic features from environmental noise

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent identifies and utilizes specific local acoustic features (acoustic feature amounts that remain constant across different environments) while ignoring other features that are sensitive to environmental changes, applying different processing qualities to different feature types

Inventive Principle:

Principle #3Local quality

2Device complexity

If traditional voice analysis methods are used, then the processing is simple, but the accuracy of discriminating between major depression and bipolar disorder is insufficient due to lack of effective biomarkers

Engineering Contradiction:

Improveanalysis method simplicityVSAvoiddisease discrimination accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent changes the parameters being analyzed by selecting specific acoustic feature amounts that have been proven to differ between major depression and bipolar disorder patients, transforming the analysis from general voice characteristics to disease-specific acoustic parameters

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12251222B2Device for estimating mental/nervous system diseases using voice

Publication Date: 2025.03.18 PST INC (NEWCO)

AI summary

It is an object to provide a device for estimating a plurality of mental/nervous system diseases by the voice analysis, the device being capable of estimating either major depression or bipolar disorder. Furthermore, there are provided an estimation device including an extraction means for extracting an acoustic feature amount that is not affected by a location where voices are acquired, and a method for operating the estimation device.