Conversation Analysis System for Speech Recognition Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice recognition error correction systems lack sufficient accuracy and efficiency due to their general approach, which does not consider the specific field or content of the conversation.
Innovation Solution
A conversation analysis system that identifies the field and summary of the conversation to ascertain easily misrecognized speech terms, using a voice analysis unit, topic ascertaining unit, topic-related correction term storage unit, and conversation correction unit to provide more accurate and prompt corrections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a general correction engine is used to correct voice recognition errors, then the system is simple and easy to operate, but the correction accuracy is insufficient
Solution Approach 1:
The system performs preliminary analysis of the conversation field and summary before correction. The topic ascertaining unit determines the conversation field and generates a summary in advance, which then guides the selection of appropriate correction terms from the storage unit, improving accuracy without requiring complex real-time processing
Solution Approach 2:
The patent introduces an intermediary structure (topic-related correction term storage unit) that stores pre-prepared correction terms organized by conversation field and summary. This intermediary layer bridges the simple correction engine and the need for accurate, context-specific corrections, allowing the system to maintain operational simplicity while achieving high correction accuracy
2Measurement precision
If field-specific correction terms are stored and used, then correction accuracy improves, but the storage and processing complexity increases
Solution Approach 1:
The correction term storage is segmented by conversation field and summary categories. Each segment contains correction terms specific to that field-summary combination, allowing the system to retrieve only relevant corrections without processing the entire database, thus managing complexity through structured organization
Solution Approach 2:
Different correction terms are stored for different conversation fields and summaries, ensuring that each local segment (field-summary combination) has its own specialized correction vocabulary. This local quality approach allows high accuracy for each specific context while maintaining an organized, manageable overall structure
3Measurement precision
If the conversation topic is analyzed before correction, then correction precision improves, but the processing time increases
Solution Approach 1:
The topic analysis (field determination and summary generation) is performed as a preliminary step before correction. By establishing the conversation context in advance, the system can quickly retrieve appropriate correction terms without performing complex analysis during the correction phase, thereby improving precision while managing processing time
Solution Approach 2:
The system uses the conversation field and summary as feedback to guide the correction process. The topic ascertaining unit provides feedback about the conversation context, which then directs the correction unit to select appropriate terms from the storage unit, creating an efficient feedback loop that improves precision without excessive time loss
Data Source
AI summary
To provide a system capable of correcting an error in speech recognition by easier work than in the prior art. A conversation analysis system comprising: a speech analysis unit 3 for analyzing contents contained in a conversation; a topics grasping unit 5 for grasping topics of the conversation; a topics-related correction terms storage unit 7 for, with respect to the respective topics, storing pronunciations to be speech-recognized or speech-recognized terms, and candidates for correction terms related to the pronunciations to be speech-recognized or the speech-recognized terms; and a conversation correction unit 9 for correcting the conversation analyzed by the speech analysis unit 3.