Speech Recognition Control for Telephone Call Resource Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech recognition systems waste resources by performing real-time speech recognition even when the recognized outcomes are not referenced, such as when an operator does not activate the user interface for checking the outcomes.
Innovation Solution
A speech recognition system that includes a control part to determine whether to execute real-time speech recognition on voice data from telephone calls, a recognition part to perform speech recognition when determined, and a UI providing part to display the recognized text on a terminal, with real-time speech recognition only executed when the text is referenced.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If real-time speech recognition is applied to all telephone calls, then the recognized text can be referenced in real-time on the terminal screen, but CPU resources are wasted when the screen is not displayed or not referenced
Solution Approach 1:
The system dynamically adjusts the speech recognition execution mode (real-time vs. non-real-time) based on whether the terminal screen is currently displayed. When the screen is displayed, real-time speech recognition is executed; when not displayed, non-real-time speech recognition is executed. This dynamic adaptation resolves the contradiction by matching resource consumption to actual user needs.
Solution Approach 2:
The system changes the operational parameter of speech recognition from a fixed real-time mode to a flexible mode that switches between real-time and non-real-time execution. This parameter change allows the system to optimize CPU resource usage while maintaining the capability for real-time recognition when needed.
2Reliability
If speech recognition is executed on all telephone calls, then complete text coverage is achieved, but resource consumption increases unnecessarily
Solution Approach 1:
The system applies different speech recognition execution qualities to different telephone calls based on local conditions (whether the terminal screen is displayed). Calls with displayed screens receive real-time recognition treatment, while calls without displayed screens receive non-real-time treatment. This local quality differentiation maintains text coverage completeness while improving overall resource utilization efficiency.
3Loss of time
If real-time speech recognition is always active, then immediate text availability is ensured, but system complexity increases
Solution Approach 1:
The system uses feedback from the terminal about whether the screen is displayed to control speech recognition execution. This feedback mechanism allows the system to automatically adjust between real-time and non-real-time modes without complex manual control logic, resolving the contradiction between immediate text availability and system complexity.
Data Source
AI summary
A speech recognition system according to one embodiment includes: a speech recognition control part configured to determine whether or not to execute speech recognition on a real-time basis, on voice data that is obtained from a voice telephone call; a speech recognition part configured to execute the speech recognition on the voice data when execution of the speech recognition on a real-time basis is determined, and generate text that represents an outcome of the speech recognition; and a UI providing part configured to display a screen, on which the generated text can be referenced on a real-time basis, on a terminal connected via a communication network, and the speech recognition control part is further configured to determine, when the screen is displayed on the terminal, executing the speech recognition on a real-time basis, on voice data being a source of text that can be referenced on the screen.