Speech Recognition System for Intent-Based Query Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing display apparatuses face challenges in accurately determining user intentions from non-sentence speech voice, leading to inconveniences in providing relevant search results due to noise or mismatch with preset patterns.
Innovation Solution
A display apparatus with an input unit, communication unit, and processor that receives user speech voice, creates and displays question sentences by comparing similarity in pronunciation between stored keywords and spoken words, and transmits these to an answer server for accurate answer retrieval, using natural language processing to extract object names and combine keywords for improved query generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a keyword recognition method is used to perform search based on core keyword from speech voice, then the display apparatus can provide search results, but users experience inconvenience by having to search for desired information from numerous search results
Solution Approach 1:
The speech recognition approach is segmented into two distinct methods: keyword recognition for simple queries and sentence recognition for complex questions. This segmentation allows the system to handle different types of user inputs appropriately, providing direct answers for sentence-based questions rather than overwhelming users with numerous keyword search results
Solution Approach 2:
The system dynamically selects between keyword recognition and sentence recognition methods based on the characteristics of the input speech. When the speech matches a preset pattern, sentence recognition is applied to provide direct answers; otherwise, keyword recognition is used. This dynamic adaptation improves both search efficiency and user convenience
2Measurement precision
If a sentence recognition method is used to analyze speech voice and provide answer result, then the answer result is closer to user's speech intention, but when a sentence speech appropriate to a preset pattern is not input or noise occurs, the method does not perform correct voice recognition
Solution Approach 1:
The system stores multiple preset question patterns and keywords in advance to cushion against recognition failures. When noise occurs or the input doesn't match preset patterns, the system can still provide relevant answers by matching against the pre-stored keywords and patterns, ensuring reliable operation under various conditions
Solution Approach 2:
Keywords serve as an intermediary between sentence recognition and answer generation. When sentence recognition fails due to noise or pattern mismatch, the system extracts keywords from the speech and matches them against pre-stored keywords to generate answers, providing a fallback mechanism that maintains recognition reliability
3Quantity of substance
If the display apparatus displays numerous search results related to speech voice, then comprehensive information is provided, but users have to search through many results to find desired information
Solution Approach 1:
The system extracts only the essential answer from the search results based on sentence recognition analysis, rather than displaying all search results. By taking out the most relevant information directly, the system provides comprehensive information coverage while eliminating the need for users to search through numerous results, thus reducing time loss
Data Source
AI summary
A display apparatus and a method for questions and answers includes a display unit includes an input unit configured to receive user's speech voice; a communication unit configured to perform data communication with an answer server; and a processor configured to create and display one or more question sentences using the speech voice in response to the speech voice being a word speech, create a question language corresponding to the question sentence selected from among the displayed one or more question sentences, transmit the created question language to the answer server via the communication unit, and, in response to one or more answer results related to the question language being received from the answer server, display the received one or more answer results. Accordingly, the display apparatus may provide an answer result appropriate to a user's question intention although a non-sentence speech is input.