Dual Voice Recognition Engine for Mobile Terminal Security
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile terminals face challenges in enhancing the reliability and security of voice recognition functions, particularly in ensuring the accuracy and speed of voice recognition results while preventing the leakage of personal information.
Innovation Solution
A mobile terminal with dual voice recognition engines, one on the server and one locally, processes user voice inputs, calculates similarity between recognition data, and selects the most reliable data for processing, with the option to block personal information from being transmitted to the server, ensuring secure and accurate voice recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If voice recognition is performed using only a local voice recognition engine, then processing speed is improved, but recognition accuracy deteriorates
Solution Approach 1:
The voice recognition system is segmented into two independent engines: a local voice recognition engine for rapid processing and a server-side voice recognition engine for high-accuracy processing. Each engine operates independently and can be selected based on the specific requirements of the task, thus resolving the contradiction between speed and accuracy.
Solution Approach 2:
The system dynamically selects between local and server-side voice recognition engines based on real-time conditions such as network availability, computational resources, and recognition requirements. This dynamic adaptation allows the system to optimize for speed when appropriate and for accuracy when needed, resolving the speed-accuracy tradeoff.
2Measurement precision
If voice recognition results from server and mobile device are combined, then recognition accuracy is improved, but information security deteriorates
Solution Approach 1:
The system extracts and processes personal information locally using the local voice recognition engine, preventing sensitive data from being transmitted to the server. Only non-sensitive voice data is sent to the server for recognition, thus maintaining accuracy while protecting personal information security.
Solution Approach 2:
Different processing locations are assigned different functions: the local device handles sensitive personal information recognition with high security requirements, while the server handles general voice recognition tasks. This local quality differentiation resolves the contradiction between accuracy and security.
3Adaptability or versatility
If distributed voice recognition system is implemented, then resource utilization is improved, but system complexity deteriorates
Solution Approach 1:
The local voice recognition engine is designed to be universal, capable of handling both simple and complex voice recognition tasks independently. This multi-functionality reduces the need for complex coordination with the server, simplifying the overall system architecture while maintaining resource distribution capabilities.
Solution Approach 2:
The system performs preliminary voice processing and filtering locally before transmitting to the server, reducing the complexity of server-client coordination. This preliminary action prepares the data in advance, making the distributed system easier to manage and less complex.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
The present disclosure relates to a mobile terminal and a voice recognition method thereof. An embodiment of the present disclosure relates to a voice recognition method of a mobile terminal in connection with a server. The voice recognition method may include receiving a user's voice; providing the received voice to a first voice recognition engine provided in the server and a second voice recognition engine provided in the mobile terminal; acquiring first voice recognition data as a result of recognizing the received voice by the first voice recognition engine; acquiring second voice recognition data as a result of recognizing the received voice by the second voice recognition engine; estimating a function corresponding to the user's intention based on at least one of the first and the second voice recognition data; calculating a similarity between the first and the second voice recognition data when personal information is required for the estimated function; and selecting either one of the first and the second voice recognition data based on the calculated similarity.