Empathy degree prediction system, emmpathy degree prediction method, and program
The empathy prediction system enhances accuracy by combining semantic and visual cues from speech and video data to predict empathy levels, addressing the limitations of existing methods in evaluating diverse conversational information.
WO2026116374A1PCT designated stage Publication Date: 2026-06-04HONDA MOTOR CO LTD
View PDF 4 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- HONDA MOTOR CO LTD
- Filing Date
- 2025-11-26
- Publication Date
- 2026-06-04
AI Technical Summary
Technical Problem
Existing methods for predicting empathy levels in conversations are not accurate due to the lack of comprehensive evaluation of diverse information types, such as speech content, facial expressions, and actions, which limits the precision of empathy assessment.
Method used
An empathy prediction system that combines text and video information by extracting semantic and visual cues, using a coupling unit to generate combined information, and employing a pattern-based classifier to predict empathy levels accurately.
Benefits of technology
Improves the accuracy of empathy level prediction by integrating semantic cues from speech and visual cues, such as facial expressions and head posture, enhancing the precision of empathy assessment in human-robot interactions.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure JP2025041200_04062026_PF_FP_ABST
Abstract
This empathy degree prediction system comprises: a text information acquisition unit for acquiring text information based on voice of a subject; a video information acquisition unit for acquiring video information of the subject; a semantic clue extraction unit for extracting a semantic clue from the text information; a visual clue extraction unit for extracting a visual clue from the video information; a connection unit for connecting the semantic clue and the visual clue to generate connection information; and an empathy degree prediction unit for predicting an empathy degree of the subject on the basis of the connection information.
Need to check novelty before this filing date? Find Prior Art