Speech Control Search Range Segmentation for Terminal Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech control technologies on terminal devices suffer from low efficiency and accuracy in responding to user instructions, leading to issues like error responses, repeated responses, and lack of response, due to the complexity of systems with multiple applications and interfaces.
Innovation Solution
A speech control method that receives and recognizes user instructions, searches for an execution object within a preset search range that includes the current interface, applications, and the system, using semantic information and priority orders to determine the appropriate response, thereby improving relevance and accuracy of responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the search range for speech recognition is expanded to cover the entire system, then the completeness of response is improved, but the processing time and computational complexity increase
Solution Approach 1:
The search range is segmented into three distinct levels: current interface elements, running applications, and system-level functions. This hierarchical segmentation allows the system to search through scopes in an organized manner, improving both completeness and efficiency by systematically covering all possible execution objects without overwhelming computational burden.
Solution Approach 2:
The system performs preliminary actions by first searching within the current interface scope before expanding to applications and system level. This preliminary search approach allows the system to quickly resolve common commands that are visible on the current screen, reducing average processing time while maintaining complete coverage for more complex commands.
2Measurement precision
If multiple search scopes are introduced to improve response accuracy, then the precision of execution object identification is improved, but the device complexity increases
Solution Approach 1:
The patent introduces a dimensional hierarchy for search scopes, organizing execution objects across three levels: interface dimension (current screen elements), application dimension (running applications), and system dimension (system-level functions). This dimensional organization improves identification accuracy by providing structured search paths while managing complexity through clear hierarchical boundaries.
Solution Approach 2:
The current interface acts as an intermediary layer between the user and the underlying applications/system. By first searching within this intermediary interface scope, the system can resolve many commands without directly accessing complex application or system structures, thereby improving accuracy while shielding the system from excessive complexity.
3Productivity
If the search range is limited to current interface only, then the processing speed is improved, but the coverage of executable commands is reduced
Solution Approach 1:
The search scope is made dynamic rather than static. The system adaptively expands or contracts the search range based on the specific command being processed. For interface-level commands, only the current interface is searched (fast). For application-level commands, the search dynamically includes running applications. For system-level commands, the full system scope is activated. This dynamic approach optimizes both speed and coverage.
Solution Approach 2:
The search scopes are nested in a hierarchical structure where the current interface scope is nested within the application scope, which is in turn nested within the system scope. This nested arrangement allows the system to start searching from the innermost (smallest) scope and progressively expand outward only when necessary, maintaining high processing speed for simple commands while ensuring comprehensive coverage for complex commands.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A speech control method, for a terminal device, includes: receiving an input speech control instruction, obtaining a recognition result of the speech control instruction; searching for an execution object matching the recognition result step by step within a preset search range; and responding to the speech control instruction based on a search result; in which the preset search range at least includes any one of: a current interface of the terminal device when receiving the speech control instruction, at least one application currently running on the terminal device when receiving the speech control instruction, and a system of the terminal device.