Automated Voice Testing Tool for Hands-Free ECU Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated testing techniques for voice-based systems, such as hands-free systems in vehicles, lack efficiency and sophistication, particularly in simulating and evaluating voice commands across various languages, accents, volumes, and speeds, which complicates the validation of electronic control units (ECUs) connected via control area networks (CAN buses.
Innovation Solution
An automated testing tool that generates simulated audible voice commands based on text-to-speech logic, communicates these commands through a CAN bus, and compares expected responses to actual outputs, enabling verdict determination, and integrates with mobile phones for testing voice-controlled actions like calls and screen changes, using a CAN message processor for communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional automated testing techniques are used for voice-based systems, then testing can be performed, but the testing efficiency and sophistication are insufficient
Solution Approach 1:
The patent introduces a voice command generator as an intermediary component that automatically generates diverse voice commands (varying in language, accent, volume, speed) and transmits them to the hands-free system under test. This mediator handles the complexity of voice parameter generation, allowing the testing framework to achieve high sophistication without increasing overall system complexity.
Solution Approach 2:
The patent creates virtual copies of human voice interactions by generating synthetic voice commands that replicate various linguistic characteristics (different languages, accents, volumes, speeds). These copied voice patterns enable comprehensive testing without requiring actual human testers, thereby improving productivity while maintaining testing sophistication.
2Adaptability or versatility
If comprehensive voice command testing is performed across various languages, accents, volumes, and speeds, then testing coverage is improved, but the complexity of the testing system increases
Solution Approach 1:
The patent segments the voice command generation process into independent controllable parameters (language, accent, volume, speed). Each parameter can be adjusted separately through the CAN bus interface, allowing comprehensive testing coverage without creating a monolithic complex system. The segmentation enables modular configuration of test scenarios.
Solution Approach 2:
The patent systematically varies multiple voice parameters (language type, accent characteristics, volume levels, speech speed) to create diverse test cases. By changing these parameters independently through the automated testing tool, the system achieves high adaptability and versatility while managing complexity through parameterized control rather than structural complexity.
3Measurement precision
If manual testing of voice commands is performed, then detailed evaluation is possible, but time consumption increases
Solution Approach 1:
The patent implements self-service automated testing where the testing system automatically generates voice commands, transmits them via CAN bus to the hands-free system, receives responses, and evaluates results without human intervention. The system performs detailed evaluation of voice command recognition accuracy, response correctness, and system behavior, achieving measurement precision equivalent to manual testing while eliminating time consumption.
Solution Approach 2:
The patent establishes a feedback loop where the automated testing tool sends voice commands to the hands-free system, receives the system's responses through the CAN bus, and automatically compares actual responses with expected outcomes. This closed-loop feedback mechanism enables detailed evaluation precision with automated time-efficient processing, eliminating the need for manual testing cycles.
Data Source
AI summary
An electronic control unit (ECU) of a hands-free module may be tested by an automated voice based testing tool in a first device. The tool reads test input data from an Excel input file. The tool generates simulated audible voice commands in a specified language, accent, pitch, volume or speed to test the hands-free module. The voice commands are transmitted via a speaker to a hands-free module microphone. The hands-free ECU is coupled to a CAN bus and the tool receives CAN bus information corresponding to hands-free module operations. The tool outputs test verdict information and/or CAN bus message logs as text in an Excel file.