Vehicle Voice Recognition Canceling Incorrect Commands
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice recognition systems in vehicles often cause inconvenience when users utter incorrect commands, as they fail to recognize negative interjections, leading to unwanted information retrieval or failure to retrieve desired information.
Innovation Solution
A voice recognition device and method that detects a negative interjection by analyzing background energy and recognizing a second voice signal after the initial command, allowing users to cancel the input command and reutter a new one, using a control device with an input device, storage device, and output device to manage voice signals and contexts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the voice recognition system detects the end point of the first voice signal and closes the input device, then the system can process the command efficiently, but the user cannot cancel the command by uttering a negative interjection
Solution Approach 1:
The system performs preliminary action by detecting the end point of the first voice signal and preparing to close the input device, but delays the actual closing action to allow time for detecting negative interjections. This resolves the contradiction by maintaining productivity through efficient processing while enabling adaptability through the extended detection window for cancellation commands.
2Adaptability or versatility
If the system keeps the input device open to allow command cancellation, then the user can utter negative interjections, but the system cannot efficiently process commands and may retrieve unwanted information
Solution Approach 1:
The system applies dynamics by making the input device state flexible rather than fixed. After detecting the end point of the first voice signal, the system dynamically extends the detection period to monitor for negative interjections, then closes the input device based on whether cancellation is detected. This resolves the contradiction by adapting the device state based on real-time conditions, enabling both efficiency and cancellation capability.
3Device complexity
If the voice recognition system recognizes only the first voice signal as a command, then the recognition process is simple and fast, but the system cannot distinguish between correct commands and corrections
Solution Approach 1:
The system segments the voice input processing into distinct phases: first voice signal detection, end point detection, negative interjection detection period, and final command execution. By dividing the recognition process into these segments, the system maintains relative simplicity while improving reliability through the additional verification step of detecting cancellation commands before executing the original command.
Data Source
AI summary
A voice recognition device for a vehicle includes: an input device receiving a command and a negative interjection uttered by a user, converting the command into a first voice signal, and converting the negative interjection into a second voice signal; a storage device storing a negative context, an interjection context, and an acoustic model; and a control device receiving the first voice signal, detecting a first start point and a first end point of the first voice signal, receiving the second voice signal after the detection of the first start point and the first end point of the first voice signal, detecting a second start point and a second end point of the second voice signal, and recognizing the second voice signal based on at least one of the negative context, the interjection context, and the acoustic model when the reception of the first voice signal and the second voice signal is completed.