Voice Assistant Response Reordering by Domain Priority
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice assistants operate by providing one response to one user utterance, failing to effectively manage multiple intents and domains in natural language understanding, leading to inefficient response ordering and user experience.
Innovation Solution
An electronic device with a processor and memory that extracts multiple intents from user input and rearranges responses based on domain information, grouping similar intents together and prioritizing them for a more coherent and natural output order.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional voice assistants provide one response to one user utterance in the order of reception, then the system operation is simple, but the response order becomes inefficient and lacks coherence when multiple intents are present
Solution Approach 1:
The patent segments the response processing into distinct phases: extracting multiple intents from user input, determining domains for each intent, and separately ordering responses based on domain priorities. This segmentation allows complex multi-intent handling while maintaining manageable system architecture.
Solution Approach 2:
The system performs preliminary classification of intents into domains before generating responses. By pre-determining domain assignments and establishing domain priority orders in advance, the system prepares the response ordering framework before actual response generation, improving overall processing efficiency.
2Ease of operation
If responses are provided in the order of utterance reception, then the processing logic is simple, but the conversation flow and context relevance deteriorate
Solution Approach 1:
The patent changes the ordering parameter from temporal sequence (order of reception) to semantic categorization (domain-based priority). By transforming the sorting criterion from time-based to meaning-based, the system maintains operational simplicity while significantly improving context coherence and conversation flow.
Solution Approach 2:
The patent introduces domain classification as an intermediary layer between intent extraction and response generation. This intermediary categorization system acts as a mediator that organizes multiple intents into coherent domain groups, ensuring context relevance without complicating the overall processing logic.
3Device complexity
If multiple intents are processed without domain-based ordering, then the system structure remains simple, but the user experience and response coherence worsen
Solution Approach 1:
The patent implements a universal domain classification framework that can handle multiple intents across various domains (weather, news, sports, etc.) using a single standardized approach. This multi-functional domain system improves user experience quality without requiring separate processing structures for different intent types.
Data Source
AI summary
Disclosed are an electronic device and a response processing method for a user of the electronic device. The electronic device includes a processor and a memory storing instructions. When the instructions are executed by the processor, the instructions may cause the electronic device to receive a user input. When the instructions are executed by the processor, the instructions may cause the electronic device to extract a first intent, a second intent, and a third intent from user input in order of reception of utterances. When the instructions are executed by the processor, the instructions may cause the electronic device to output a first response, a second response, and a third response respectively corresponding to the first intent, the second intent, and the third intent by arranging the response in a different order from the order of reception of utterances based on respective domains of the first intent, the second intent, and the third intent.