Vehicle Voice Agent Response Disambiguation by Location Cues

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle agent processing devices struggle to formulate suitable responses when users pose ambiguous questions, as they rely on clear wording to provide accurate information about vehicle operations.

Innovation Solution

The agent device includes a recognition section to identify spoken content and an output section that generates a relevant list of responses when clear information is not available, using predetermined words related to vehicle functions, locations, shapes, or colors, and outputs pre-prepared candidate images or error messages to assist users in finding suitable responses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the agent processing device relies on clear wording to provide accurate information, then the precision of information provided is improved, but the adaptability to handle ambiguous questions deteriorates

Engineering Contradiction:
Improveprecision of information providedVSAvoidadaptability to handle ambiguous questions
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary actions by pre-generating multiple candidate responses and storing them in advance. When an ambiguous question is detected, the system retrieves and presents these pre-prepared candidate responses instead of requiring clear wording, thus maintaining adaptability while preserving information precision through user selection.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediary mechanism by presenting multiple candidate responses as intermediate options between the ambiguous question and the final accurate information. This intermediary step allows the system to handle ambiguous input while still providing precise information through the user's selection from the candidates.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If the system outputs a relevant list for ambiguous questions, then the adaptability to handle various user inputs is improved, but the device complexity increases

Engineering Contradiction:
Improveability to handle ambiguous questionsVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs preliminary action by pre-generating and storing multiple candidate responses for potential ambiguous questions. This advance preparation enables the system to quickly retrieve and present relevant candidates without complex real-time analysis, reducing operational complexity while maintaining high adaptability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system employs self-service by automatically generating and presenting candidate responses without requiring complex external processing. The agent processing device independently handles ambiguous questions by retrieving pre-prepared candidates and presenting them to the user, simplifying the overall system architecture.

Inventive Principle:
Principle #25Self-service

3Reliability

If multiple candidate responses are pre-prepared and presented, then the probability of formulating suitable responses is improved, but the loss of time in processing increases

Engineering Contradiction:
Improveprobability of formulating suitable responsesVSAvoidtime for processing and presenting candidates
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-generating and storing multiple candidate responses before they are needed. When an ambiguous question is detected, the system simply retrieves and presents these pre-prepared candidates, significantly reducing processing time compared to generating responses in real-time, thus improving reliability without excessive time loss.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11869488B2Agent device, agent system, and computer-readable storage medium
Publication Date: 2024.01.09 TOYOTA JIDOSHA KK
  • US11869488B2 patent drawing
  • US11869488B2 patent drawing
  • US11869488B2 patent drawing

AI summary

In cases in which a confidence score of an inferred intent label is a predetermined threshold or less, an intent inference section searches for whether or not wording pertaining to a location, such as “on the door”, is present in a question. In cases in which a word relating to a location is present, the intent inference section consults individual function identification data associated with OM item codes in order to find intent labels including individual functions relevant to the location (such as “door”). In cases in which an intent label including an individual function relevant to the “door” is found, an OMA interaction control section consults QA data to find and acquire associated response information based on the found intent label and the OM item code, and notifies a HMI interaction control section of such response information.