Confirmation Response Generation Using Dictionary Data to Reduce Misunderstanding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing conversation systems face challenges in accurately interpreting user speech due to low reliability of voice recognition and intention estimation, leading to misunderstandings when multiple interpretations exist, and fail to correct mishearings effectively.
Innovation Solution
An information processing device and system that includes an audio input unit, a confirmation response generating unit, and an audio output unit, which uses additional information from dictionary data and application execution results to generate confirmation responses, reducing the likelihood of misunderstandings by providing clearer responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a confirmation response is generated by repeating the user's speech without change, then the system can request confirmation of user intent, but the user may mishear or misunderstand the response, leading to incorrect information being provided
Solution Approach 1:
The patent introduces an intermediary process between voice recognition and confirmation response generation. Instead of directly repeating the recognized speech, the system uses additional information from dictionary data and application execution results to mediate the response generation, thereby reducing mishearing and misunderstanding while maintaining confirmation accuracy
Solution Approach 2:
The system implements feedback by using application execution results and dictionary data to verify and refine the recognition results before generating confirmation responses. This feedback loop ensures that the confirmation response contains accurate information that can help the user correct mishearing or misunderstanding
2Reliability
If additional information from dictionary data and application execution results is used to generate confirmation responses, then misunderstanding is reduced, but system complexity increases
Solution Approach 1:
The patent makes the confirmation response generating unit multi-functional by enabling it to not only repeat user speech but also to access and process additional information from dictionary data and application execution results. This universal approach allows the same unit to handle both simple repetition and complex information verification without adding separate dedicated components
Solution Approach 2:
The system merges the voice recognition function, additional information processing function, and confirmation response generation function into an integrated confirmation response generating unit. This combination reduces overall system complexity by eliminating the need for separate dedicated components for each function
Data Source
AI summary
When a confirmation response to a user speech is generated, a confirmation response capable of reducing a possibility of the occurrence of misunderstanding by adding additional information acquired from dictionary data or an application execution results is generated. Provided is an information processing device including an audio input unit that receives a user speech, a confirmation response generating unit that generates a confirmation response for requesting confirmation of a speaking user to the user speech, and an audio output unit that outputs the confirmation response generated by the confirmation response generating unit, in which the confirmation response generating unit executes generation of a confirmation response using additional information acquired from at least any of dictionary data and an application execution result. For example, the confirmation response is generated using general knowledge base dictionary data, an application execution result, or information obtained from a captured image, a sensor, or the like.