Context Information Manager for NLU Modules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices face inconvenience in sharing context information necessary for multiple services, leading to duplicative user interactions due to independent analysis systems that cannot share information across different modules.
Innovation Solution
A system comprising a communication interface, processor, and memory that includes an automatic speech recognition (ASR) module and multiple natural language understanding (NLU) modules, allowing for the processing of user utterances, generation of intent and context information, and sharing of context information between modules for integrated service processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If independent NLU modules are used for different services, then service specialization and accuracy are improved, but context information cannot be shared between modules
Solution Approach 1:
The patent introduces a context information manager as an intermediary component that collects, manages, and distributes context information extracted from user utterances to multiple NLU modules. This mediator enables information sharing between independent service modules without compromising their specialized processing capabilities, resolving the contradiction between module independence and information sharing.
2Adaptability or versatility
If multiple independent analysis systems are deployed for various services, then service coverage and functionality are improved, but duplicative user interactions increase
Solution Approach 1:
The patent merges the context information processing functionality across multiple independent NLU modules by introducing a shared context information manager. This allows different service modules to access and utilize the same context information extracted from user utterances, eliminating the need for duplicative interactions while maintaining comprehensive service coverage.
3Device complexity
If separate modules process user utterances for different services, then service independence and modularity are improved, but information reuse across services decreases
Solution Approach 1:
The patent segments the information processing architecture into specialized NLU modules for different services while introducing a separate context information manager component. This segmentation maintains service independence and modularity while enabling efficient information reuse through the dedicated context management component that serves all modules.
Data Source
AI summary
A system is provided. The system may include a communication interface, at least one processor, and at least one memory. The memory may store instructions that, when executed, cause the processor to provide an ASR module and a plurality of NLU modules, and further cause the processor to receive first voice data including a first user utterance, to process the first voice data using the ASR module to generate first text data, to process the first text data using a first NLU module into a first intent and one or more context information items associated with the first intent, to provide a first response based on the first intent and the one or more context information items, and to provide at least part of the one or more context information items to at least one other NLU module.