Conversation Content Provision via Voice Segmentation and Ontology Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice recognition technologies face difficulties in understanding conversations between users and others due to issues like pronunciation, environmental noise, and processing natural language, making it challenging to actively provide relevant contents or services.
Innovation Solution
A method and apparatus that collect voice information from conversations, create search keywords using morphemes and ontology, and search for relevant contents using a search engine, allowing for the provision of contents to users based on their conversation topics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If voice recognition technology is used to understand conversation, then voice-based interaction is enabled, but understanding accuracy deteriorates due to pronunciation, noise, and natural language processing difficulties
Solution Approach 1:
The patent segments the continuous voice signal into discrete sentence units using end point detection technology. This segmentation allows the system to process and analyze individual sentences separately, improving accuracy by focusing on complete semantic units rather than continuous speech streams, thereby addressing the precision problem while maintaining voice-based interaction ease.
Solution Approach 2:
The patent introduces an intermediary processing layer that includes morpheme extraction, subject word identification, and ontology-based mapping. This intermediary layer transforms raw voice data into structured semantic representations, bridging the gap between noisy voice input and accurate conversation understanding, thus resolving the contradiction between ease of voice interaction and understanding precision.
2Measurement precision
If complex voice processing is applied to improve conversation understanding, then accuracy improves, but processing time and system complexity increase
Solution Approach 1:
By segmenting voice input into sentence units, the system breaks down complex processing into manageable discrete units. This segmentation reduces the overall processing complexity while maintaining accuracy, as each sentence unit can be processed independently through the morpheme extraction and ontology mapping pipeline.
Solution Approach 2:
The patent applies partial action by focusing processing efforts on key elements such as subject words and morphemes rather than analyzing every word equally. The ontology-based mapping selectively processes only the most relevant semantic elements, reducing unnecessary processing complexity while preserving understanding accuracy.
3Measurement precision
If end point detection and sentence division are used to process voice information, then processing precision improves, but processing time increases
Solution Approach 1:
The patent performs preliminary end point detection and sentence division before the main morpheme extraction and ontology mapping processes. By pre-segmenting the voice input into sentence units, the system prepares the data structure in advance, allowing subsequent processing steps to work on pre-organized units, which reduces overall processing time while maintaining detection precision.
Data Source
AI summary
Disclosed are a method and an apparatus for providing contents about conversation, which collect voice information from conversation between a user and another person, search contents on the basis of the collected voice information, and provide contents about the conversation between the user and the person. The method of providing contents about conversation includes: a voice information collecting step of collecting voice information from conversation between a user and another person; a keyword creating control step of creating search keywords by using the collected voice information; and a contents providing control step of searching contents by using the created search keywords, and providing the searched contents.