Voice Assistant Anaphor Resolution Using Common Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice assistant systems struggle to accurately interpret user intentions when anaphors are included in user inputs, making it difficult to provide specialized services considering the meaning of anaphors.
Innovation Solution
A system and method that utilizes common information representing target words to interpret anaphors within text, using a non-transitory computer-readable recording medium to execute operations that generate and manage common information for voice assistant services, enabling accurate identification and response to user inputs containing anaphors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional voice recognition technology is used, then basic voice input processing is achieved, but accurate interpretation of anaphors and user intentions cannot be accomplished
Solution Approach 1:
The system performs preliminary action by generating and storing common information about target words (such as entities, locations, or objects) before the user's query is processed. This common information is retrieved and used to resolve anaphors in the user's input, enabling accurate interpretation without requiring complex real-time analysis of the anaphor resolution itself.
2Adaptability or versatility
If voice assistant services are specialized for multiple domains, then service versatility is improved, but difficulty in considering anaphor meaning across domains increases
Solution Approach 1:
The system implements universality by creating a common information management mechanism that serves multiple domain-specific voice assistant services. The common information (entities, locations, objects) is stored in a centralized structure and can be retrieved and applied across different domains, allowing each domain-specific service to leverage shared knowledge without requiring domain-specific anaphor resolution logic.
3Measurement precision
If common information about target words is generated and managed, then anaphor interpretation accuracy is improved, but system complexity increases
Solution Approach 1:
The system introduces an intermediary mechanism (common information management module) that sits between the user input processing and the domain-specific voice assistant services. This intermediary generates, stores, and manages common information about target words, acting as a mediator that translates raw user input into structured knowledge that can be efficiently utilized by multiple services without adding complexity to each individual service.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for providing a voice assistant service for text including an anaphor are provided. A method, performed by an electronic device, of providing a voice assistant service includes: obtaining first text generated from a first input, detecting a target word within the first text and generating common information related to the detected target word, using a first natural language understanding (NLU) model, obtaining second text generated from a second input, inputting the common information and the second text to a second NLU model, detecting an anaphor included in the second text and outputting an intent and a parameter, based on common information corresponding to the detected anaphor, using the second NLU model, and generating response information related to the intent and the parameter.