Real-Time Call Transcript Display for Reviewing Missed Voice Content
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users often miss portions of voice calls due to interruptions or multitasking, making it difficult to check past call content in real-time environments.
Innovation Solution
Converting real-time call voice into text and displaying it with speaker information, allowing users to review past call content through user inputs and screen switching, including picture-in-picture formats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If real-time voice call is conducted, then communication efficiency is improved, but user may miss call content due to multitasking or interruptions
Solution Approach 1:
The patent creates a visual copy of the voice call content by converting speech to text and displaying it on the screen. This text copy allows users to review call content visually without missing any information, even when multitasking or interrupted during the real-time call.
Solution Approach 2:
The patent performs preliminary conversion of voice to text during the call and prepares the text display in advance. This allows the call content to be readily available for review when users need to check missed information, eliminating the need to wait until after the call ends.
2Loss of information
If user checks past call content during real-time call, then information completeness is improved, but call flow is interrupted
Solution Approach 1:
The patent implements dynamic control of the text display, allowing users to toggle between viewing past call content and maintaining real-time call flow. The system adapts to user needs by enabling easy switching between reviewing historical text and continuing the live conversation without permanent interruption.
Solution Approach 2:
The patent allows users to review only the specific portion of call content they need to check, rather than requiring them to review the entire call transcript. This partial review approach minimizes disruption to the ongoing call while still providing access to necessary information.
3Ease of operation
If voice call is conducted without visual display, then call simplicity is maintained, but user cannot review past content easily
Solution Approach 1:
The patent creates a visual text copy of the voice call that users can review at any time. This text representation preserves the simplicity of voice calling while adding the capability to visually inspect past content, combining the benefits of both voice and text communication modes.
Solution Approach 2:
The patent makes the call system multi-functional by integrating both real-time voice communication and visual text review capabilities into a single interface. This allows the system to serve multiple purposes: active conversation and historical review, without requiring separate applications or complex operations.
4Loss of information
If text conversion and display is added to voice call, then call content reviewability is improved, but system complexity increases
Solution Approach 1:
The patent introduces text conversion as an intermediary process between voice input and user review. This intermediary layer translates speech to text automatically, providing a bridge that enables content review without requiring complex user operations or manual transcription processes.
Data Source
AI summary
A call voice data processing method performed by at least one processor, the call voice data processing method including converting first voice data into first text, the first voice data corresponding to a real-time call, first displaying the first text and first speaker information on a first screen, the first speaker information corresponding to a speaker of the first voice data, and second displaying second text and second speaker information on the first screen in response to receiving a first user input, the second text corresponding to second voice data associated with a first time point, the second voice data corresponding to the real-time call, the first time point being earlier than a current time by a first duration, and the second speaker information corresponding to a speaker of the second voice data.


