Dialogue Processing System Using Speech Act Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional chatbots struggle to maintain natural conversations due to inappropriate responses and repetition, as they are limited to reacting to user inquiries without effectively achieving dialogue objectives through speech act control.
Innovation Solution
A dialogue processing system using speech act control that includes a main speech act unit for purpose-oriented utterances and a free speech act unit to process deviating utterances, allowing the system to return to the main speech act by searching for nodes capable of inducing a return path, thereby maintaining conversation flow and achieving dialogue objectives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a chatbot uses conventional response processing methods, then it can respond to user inquiries, but it responds with wrong answers or repeats the same response, making conversation unnatural
Solution Approach 1:
The patent segments the dialogue processing into two distinct units: main speech act unit for purpose-oriented utterances and free speech act unit for deviating utterances. This segmentation allows each unit to specialize in handling specific types of user inputs, improving response accuracy while maintaining conversation naturalness through appropriate routing of different utterance types.
Solution Approach 2:
The patent implements dynamic switching between main speech act processing and free speech act processing based on the type of user utterance. The system dynamically determines whether to follow the predetermined speech act sequence or handle free-form deviations, enabling adaptable response generation that maintains both accuracy and naturalness.
2Device complexity
If a chatbot is limited to responding to user inquiries, then it can maintain simple processing logic, but it cannot achieve objectives through dialogue
Solution Approach 1:
The patent divides the chatbot functionality into main speech act unit handling purpose-oriented dialogues and free speech act unit handling deviating utterances. This segmentation enables the system to pursue dialogue objectives through structured main speech acts while accommodating natural conversation deviations through free speech acts, increasing versatility without overwhelming complexity.
Solution Approach 2:
The patent introduces a speech act analysis unit as an intermediary that analyzes user utterances and determines whether they are purpose-oriented or deviating. This intermediary component enables the system to achieve dialogue objectives by routing appropriate utterances to the main speech act unit while maintaining conversational flexibility through the free speech act unit.
3Productivity
If a chatbot follows a predetermined speech act sequence, then it can achieve dialogue objectives, but it cannot handle user deviations naturally
Solution Approach 1:
The patent segments speech act processing into main speech act unit for predetermined sequences and free speech act unit for deviations. This allows the system to efficiently pursue dialogue objectives through structured main speech acts while simultaneously handling user deviations naturally through the free speech act unit, resolving the contradiction between productivity and adaptability.
Solution Approach 2:
The patent creates a universal dialogue processing system that can handle both purpose-oriented utterances and deviating utterances through its dual-unit architecture. The main speech act unit and free speech act unit work together to achieve dialogue objectives while accommodating various user input types, making the system multi-functional and adaptable.
4Adaptability or versatility
If a chatbot processes free speech acts, then it can handle deviating utterances, but it loses track of the main speech act
Solution Approach 1:
The patent introduces a speech act analysis unit as an intermediary that monitors and analyzes user utterances to determine whether they are purpose-oriented or deviating. This intermediary ensures that even when processing free speech acts, the system maintains awareness of the main speech act context and can return to it when appropriate, stabilizing the dialogue structure.
Solution Approach 2:
The patent implements feedback mechanisms where the free speech act unit monitors the dialogue state and provides feedback to maintain context awareness. This feedback ensures that while handling deviating utterances, the system remains aware of the main speech act trajectory and can guide the conversation back to objectives when needed.
Data Source
AI summary
Disclosed is a dialogue processing system using speech act control, the dialogue processing system comprising: a main speech act unit which processes a free speech act and performs speech act control such that the free speech act returns to a main speech act, thereby processing a multi-turn dialogue in a consistent manner, and which processes a purposed utterance having a set purpose, for reaching a final dialogue objective; and a free speech act unit which processes a free utterance deviating from the purposed utterance and performs control such that the free utterance returns to the main speech act unit by searching for a node capable of returning to the purposed utterance.