Speech Command Management via Application State Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition systems in car navigation and similar systems face decreased recognition ratios when managing a large number of commands, especially in systems where applications can be dynamically installed or changed, limiting the user's ability to efficiently interact with multiple applications simultaneously through speech.
Innovation Solution
Implementing a terminal with a control unit, speech recognition engine, and memory to manage global commands, which are associated with application IDs and states, allowing the system to adjust recognizable commands based on the current application states, preventing collisions and improving recognition efficiency by using SRGF for grammar description and dynamic prioritization or database registration to manage global commands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of commands to be recognized is increased to support more applications, then the versatility of the system is improved, but the recognition ratio deteriorates
Solution Approach 1:
The patent segments the command recognition system by application state. The control unit divides the plurality of applications into groups based on their current execution states, and the speech recognition engine processes speech inputs according to the segmented state information. This segmentation allows the system to maintain high recognition ratios for each application while supporting multiple applications simultaneously.
Solution Approach 2:
The patent implements dynamic command management where the set of recognizable commands changes based on application states. The control unit dynamically adjusts which applications are recognizable at any given moment based on their execution states, allowing the system to adapt the command vocabulary size dynamically rather than maintaining a fixed large set of commands, thereby preserving recognition ratio.
2Adaptability or versatility
If global commands are managed for all possible application states, then the adaptability is improved, but the device complexity increases
Solution Approach 1:
The control unit serves multiple functions: it monitors application states, manages the recognition state of applications, and controls speech recognition processing. This multi-functionality eliminates the need for separate dedicated components for each function, reducing overall system complexity while maintaining the ability to manage global commands across all application states.
Solution Approach 2:
Applications provide their own state information to the control unit, which then uses this information to manage recognition states. The system serves itself by automatically adjusting command recognition based on application states without requiring external configuration or complex management mechanisms.
3Productivity
If the number of simultaneously executing applications is increased, then the productivity is improved, but the ease of operation deteriorates
Solution Approach 1:
The control unit continuously monitors application execution states and uses this feedback to dynamically adjust which applications are recognizable for speech input. This feedback mechanism ensures that at any moment, the system presents a manageable set of recognizable commands corresponding to the current application state, maintaining ease of operation even when multiple applications are simultaneously executing.
Data Source
AI summary
Arrangements, including a method of speech command management that changes a content of active speech commands and inactive speech commands, based on each application status in a terminal. When plural applications are active and when a user is interacting with one of the active applications, the provided speech command management makes the content of recognizable speech commands to be limited to local commands of an interacting application, and also global commands of other background applications. Also, the provided management arrangement makes the content of recognizable global speech commands based on each application status. This management arrangement reduces the number of recognizable speech commands, and thus prevents misrecognitions inducted by using many speech commands.