Nonverbal Information Generation for Communication Robots

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing techniques for generating nonverbal information for communication robots and agents require high costs due to the need for manual creation and registration of utterance and action information.

Innovation Solution

A nonverbal information generation apparatus that automates the association of voice or text information with nonverbal information using a learned model, generating time-stamped nonverbal information based on time-stamped feature quantities and expression units.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual creation and registration of utterance and nonverbal action information is performed, then the accuracy and quality of nonverbal behavior association is improved, but the cost and time required for data creation increases significantly

Engineering Contradiction:

Improveaccuracy of nonverbal behavior associationVSAvoidtime for data creation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system automatically extracts nonverbal action information from video data and associates it with utterances using machine learning models, eliminating the need for manual annotation. The apparatus performs self-service by autonomously processing video and audio inputs to generate training data and deploy nonverbal behavior generation capabilities without human intervention in the data preparation phase.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical annotation processes with automated machine learning systems. Machine learning models analyze video and audio data to automatically extract and associate nonverbal actions with utterances, substituting human labor with computational algorithms that can process data at scale without proportional increases in time or cost.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If manual creation and registration of utterance and nonverbal action information is performed, then the quality of nonverbal behavior data is improved, but the cost of data creation increases

Engineering Contradiction:

Improvequality of nonverbal behavior dataVSAvoidcost of data creation

Core Design Contradiction:

Measurement precisionVSEase of manufacture

Solution Approach 1:

The system autonomously processes video and audio inputs to generate high-quality nonverbal behavior data without manual intervention. The machine learning models automatically extract meaningful nonverbal actions and associate them with utterances, enabling the system to serve itself in data creation while maintaining quality standards through algorithmic analysis rather than human annotation.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Manual data creation processes are replaced with automated machine learning pipelines that process video and audio data. This substitution eliminates the need for human annotators while maintaining data quality through sophisticated algorithmic extraction and association methods, thereby reducing costs without sacrificing quality.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If automated generation of nonverbal information using machine learning is implemented, then the cost and time of data creation is reduced, but the complexity of the system increases

Engineering Contradiction:

Improveefficiency of nonverbal information generationVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system is divided into distinct functional modules: video processing units, audio processing units, machine learning model components, and nonverbal action extraction modules. Each segment handles specific tasks independently, allowing the complex overall system to be managed through modular architecture where complexity is distributed across separate, specialized components rather than concentrated in a single system.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12340629B2Nonverbal information generation apparatus, nonverbal information generation model learning apparatus, methods, and programs

Publication Date: 2025.06.24 NT T INC

AI summary

A nonverbal information generation apparatus includes a nonverbal information generation unit that generates time-information-stamped nonverbal information that corresponds to time-information-stamped text feature quantities on the basis of the time-information-stamped text feature quantities and a learned nonverbal information generation model. The time-information-stamped text feature quantities are configured to include feature quantities that have been extracted from text and time information representing times assigned to predetermined units of the text. The nonverbal information is information for controlling an expression unit that expresses behavior that corresponds to the text.