In-Vehicle Voice Command Learning for Personalized Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional speech recognition technologies in vehicles lack customization and usability, failing to recognize terms specific to individual drivers and only responding with guide sounds, making them inconvenient for operation confirmation.
Innovation Solution
A vehicle system that integrates speech recognition with a graphical user interface (GUI) and a learning function for customization, using a control device to transmit and analyze speech data, register user-defined commands, and control vehicle functions based on user inputs, allowing for personalized command recognition and operation confirmation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional speech recognition technology is used in vehicles, then the system can recognize predetermined speech commands, but it cannot recognize terms that are used differently for each driver, resulting in poor usability
Solution Approach 1:
The system performs preliminary speech registration to build a personalized vocabulary for each driver before actual speech recognition operations. This preliminary action stores commonly used terms and expressions specific to each user, enabling the system to adapt to individual speech patterns and improve recognition accuracy for personalized commands.
Solution Approach 2:
The system automatically learns and adapts to each driver's speech patterns through continuous interaction. By analyzing registered speech data and identifying frequently used terms, the system self-configures personalized recognition profiles without requiring manual programming, thereby improving adaptability to individual users naturally.
2Loss of information
If conventional speech recognition technology simply outputs a guide sound, then the system provides basic feedback, but it is not possible to confirm the operation of the vehicle
Solution Approach 1:
The system implements a dual-mode feedback mechanism: for recognized speech commands, it provides both guide sounds and visual display of operation results. This feedback loop allows users to confirm whether their speech input was correctly interpreted and whether the intended vehicle operation was executed, eliminating information loss about command recognition status.
Solution Approach 2:
The system transitions from single-dimensional audio feedback to multi-dimensional feedback by incorporating visual display alongside auditory guide sounds. This dimensional expansion provides redundant confirmation channels, ensuring users can verify operations through multiple senses and improving overall feedback reliability.
3Adaptability or versatility
If the system displays information about target devices for unrecognized speech commands, then the system provides more interaction options, but the device complexity increases
Solution Approach 1:
The control device serves multiple functions: it acts as both a speech recognition system and a command registration system. By integrating the capability to learn new commands directly into the existing control architecture, the system avoids adding separate complex modules while still providing flexible command customization and target device selection capabilities.
Data Source
AI summary
An embodiment vehicle includes a communication device, an output device, an input device including a microphone, and a control device connected to each. The control device is configured to transmit speech data received through the microphone to a server through the communication device, to receive analysis information of the speech data from the server through the communication device in response to the transmission of the speech data, to identify whether the speech data corresponds to a first speech command registered in the server based on the analysis information, to control the output device to output information about target devices of a control setting of the vehicle based on the speech data not corresponding to the first speech command, and to set the speech data as a second speech command for controlling at least one of the target devices based on a reception of a user input through the input device.