Call Information Display Method with Topic Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current call recording and summarization techniques in mobile devices are inefficient, requiring significant time to find specific parts of recorded conversations and often necessitate manual note-taking, with a need for improved visualization methods to quickly recall call conversations.
Innovation Solution
A method and electronic device that classify and summarize call data by extracting speech from users and call counterparts, determining conversation topics, and displaying this information alongside call histories using various visualization techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If call conversation content is stored and offered to user, then user can easily remember past call conversation, but a long time is required for user to find desired part in recorded call conversation
Solution Approach 1:
The patent extracts key information from call conversations by performing speech-to-text conversion and identifying important entities (names, locations, organizations) and topics. This extraction process creates a condensed summary that allows users to quickly find desired information without searching through the entire recorded conversation, thus resolving the contradiction between maintaining complete information and reducing search time.
Solution Approach 2:
The patent segments call conversation data into structured components including speech-to-text content, extracted entities, identified topics, and summarized key points. This segmentation organizes the information hierarchically, allowing users to navigate from high-level topic summaries to specific conversation segments, thereby reducing the time required to locate desired information while preserving the complete conversation record.
2Loss of information
If manual note-taking is performed to remember call conversation, then call details can be recorded, but significant time and effort are required
Solution Approach 1:
The patent implements an automated system that performs speech-to-text conversion, entity extraction, topic identification, and summary generation without requiring user intervention. The system serves itself by automatically processing call recordings and generating structured notes, eliminating the need for manual note-taking while preserving all call details, thus resolving the contradiction between information retention and time investment.
3Loss of information
If call data is displayed in traditional format, then call history is recorded, but various visualization techniques are needed for quick recall
Solution Approach 1:
The patent transforms call data from traditional linear text format into a multi-dimensional structured format that includes speech-to-text content, extracted entities with categories, identified topics, and summarized key points. This dimensional transformation allows the information to be accessed from multiple angles (by topic, by entity, by time), improving accessibility without requiring complex visualization techniques, as the structured data can be displayed in simple organized formats.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for displaying call information in an electronic device is provided. The method includes receiving speech from a user of the electronic device and speech from a call counterpart, extracting data from the received speech from the user and the received speech from the call counterpart, determining a topic of a call conversation, based on the extracted data, and displaying information about the determined topic of the call conversation together with a call history of the user of the electronic device.