Button-Scoped Voice Recognition Grammar for In-Car Command Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle voice recognition systems require users to know exact command wordings, limiting their effectiveness due to factors like road noise and user accent, and necessitate learning through training, trial and error, or manual exploration.
Innovation Solution
A voice recognition system with buttons on a steering wheel dedicated to specific functionalities, disabling irrelevant grammar domains to enhance command recognition, allowing users to input commands without needing precise wording.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If all grammar domains are enabled for comprehensive voice command recognition, then the system can handle multiple functionalities, but the voice recognition accuracy decreases due to noise from irrelevant grammar domains
Solution Approach 1:
The grammar domains are segmented into multiple separate domains corresponding to different functionalities (phone, media, navigation). Each button press activates only the relevant grammar domain segment, isolating it from other segments to prevent interference and improve recognition accuracy for the specific functionality.
Solution Approach 2:
Different grammar domains are enabled locally based on the active button context. When a specific button is pressed, only the grammar domain corresponding to that button's functionality is enabled, while other domains remain disabled. This local activation ensures high recognition accuracy for the intended functionality without noise from unrelated domains.
2Measurement precision
If the system requires users to know exact command wordings, then the recognition can be more precise, but the ease of operation decreases as users must learn specific commands
Solution Approach 1:
The system performs preliminary action by pre-enabling only the relevant grammar domain corresponding to the pressed button before receiving the voice command. This preliminary configuration of the recognition system allows users to speak naturally without needing to learn specific command wordings, as the appropriate grammar is already prepared and active.
Solution Approach 2:
The system serves itself by automatically selecting and enabling the appropriate grammar domain based on the button context, eliminating the need for users to manually select or learn command structures. The system adapts to user input needs automatically, improving ease of operation while maintaining precision.
Data Source
AI summary
Voice recognition (VR) systems and methods for performing VR are provided. The VR system may comprise one or more microphones, one or more speakers, and a computing device, comprising a processor, a memory, and a plurality of buttons. Each button may be associated with a VR functionality, of a plurality of VR functionalities. Each VR functionality may have one or more grammar domains associated with it. The memory may be configured to store instructions that, when executed by the processor, are configured to cause the processor to receive an input from one of the plurality of buttons, specifying a VR functionality, disable one or more grammar domains not associated with the specified VR functionality, receive an audio input, via the one or more microphones, and analyze the audio input absent the disabled one or more grammar domains to determine one or more VR commands from the audio input.