In-Vehicle Voice Dialogue Routing for Commands and Small Talk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conversational AI systems, such as ChatGPT, are not capable of operating in-vehicle devices and do not provide natural responses to utterances that are not intended for device operation, such as small talk.
Innovation Solution
A vehicular dialogue system utilizing a conversational AI that includes a voice input unit, voice recognition unit, device control unit, voice synthesis unit, and a determination unit to distinguish between operation instructions and non-operation utterances, enabling control of in-vehicle devices and natural responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a voice agent service responds only to predefined utterance data, then the system structure is simple and easy to implement, but the response becomes unnatural and cannot handle conversations like small talk
Solution Approach 1:
The system segments utterance processing into two distinct pathways: one for predefined operation commands and another for conversational AI processing. This segmentation allows each pathway to be optimized independently, maintaining system simplicity for commands while enabling natural conversation handling through the conversational AI branch.
Solution Approach 2:
The determination unit acts as an intermediary that receives the converted text and decides whether to route it to the device control unit or the conversational AI. This intermediary component enables the system to dynamically select the appropriate processing pathway based on the utterance type, resolving the contradiction between structured command handling and natural conversation.
2Ease of operation
If a conversational AI is used for natural conversations, then the response naturalness is improved, but the ability to operate in-vehicle devices is lost
Solution Approach 1:
The system implements multi-functionality by integrating both predefined command processing and conversational AI capabilities within a single unified system. The determination unit enables the system to adaptively switch between device control mode and natural conversation mode, allowing the system to perform multiple functions (command execution and conversational interaction) without compromising either capability.
3Productivity
If the system determines whether text indicates an operation instruction, then the processing capability is improved, but the system complexity increases
Solution Approach 1:
The determination function is extracted as a separate, dedicated unit that independently evaluates whether text represents an operation instruction. This extraction allows the determination logic to be developed and optimized separately from the rest of the system, improving processing capability while containing the complexity increase to a specific, manageable component rather than distributing it throughout the entire system.
Data Source
AI summary
A voice dialogue unit determines whether the text data converted by a voice recognition unit indicates an operation instruction for an in-vehicle device. The voice dialogue unit controls, in response to determining that the text data indicates the operation instruction for the in-vehicle device, the in-vehicle device according to the operation instruction. The voice dialogue unit inputs, in response to determining that the text data does not indicate the operation instruction for the in-vehicle device, text data of voice to a conversational AI as input information.