Dual ASR Engine Voice Command Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition systems in vehicles face challenges in accurately identifying voice commands due to unfamiliarity with commands, ambient noise, and limited storage and processing capabilities, leading to reduced robustness.
Innovation Solution
The implementation of a dual automatic speech recognition (ASR) engine system, comprising a first ASR engine with an acoustic and language model, and a second ASR engine with a deep neural network, in conjunction with a habits engine using pattern recognition algorithms, to enhance voice command identification by leveraging user habits and noise threshold analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single ASR engine is used for voice command recognition, then the system is simple and processing power is saved, but the accuracy and robustness of voice command identification deteriorates due to unfamiliarity with commands and ambient noise
Solution Approach 1:
The patent divides the ASR system into multiple specialized engines: a first ASR engine for initial voice command identification and a second ASR engine for handling low-confidence or noisy commands. This segmentation allows each engine to be optimized for specific scenarios, improving overall recognition accuracy while managing system complexity through modular architecture.
Solution Approach 2:
The system dynamically changes operational parameters by switching between different ASR engines based on confidence levels and noise thresholds. When the first engine's confidence falls below a threshold or noise exceeds a threshold, the system activates the second engine with different processing parameters, allowing adaptive response to varying acoustic conditions.
2Reliability
If a dual ASR engine system is implemented to improve voice command identification accuracy, then recognition robustness improves, but processing power and memory requirements increase
Solution Approach 1:
The system applies partial action by using the second, more resource-intensive ASR engine only when necessary - specifically when confidence levels are low or noise thresholds are exceeded. For clear, high-confidence commands, only the first ASR engine is used, avoiding unnecessary processing power consumption while maintaining high robustness when needed.
Solution Approach 2:
The first ASR engine serves itself as the primary processor for most commands, only escalating to the second engine when it detects uncertainty or adverse acoustic conditions. This self-service approach allows the system to manage its own resource allocation dynamically, consuming processing power only when the situation demands enhanced recognition capabilities.
3Measurement precision
If multiple ASR engines are used to handle low confidence and high noise scenarios, then voice command identification accuracy improves, but device complexity and memory requirements increase
Solution Approach 1:
The patent segments the ASR functionality into two distinct engines with different memory footprints and processing characteristics. The first engine handles routine, high-confidence commands with lower memory requirements, while the second engine is reserved for complex, low-confidence scenarios. This segmentation optimizes memory usage by loading and executing only the necessary engine for each specific recognition task.
Data Source
AI summary
Method and apparatus are disclosed for speech recognition for vehicle voice commands. An example vehicle includes a microphone to collect a signal including a voice command, memory, and a controller. The controller is configured to determine an initial identification by feeding the signal into a first automatic speech recognition (ASR) engine and determine habits by feeding user history into a habits engine. The controller also is configured to identify the voice command by feeding the signal, the initial identification, and the habits into a second ASR engine. The controller also is configured to perform a vehicle function based on the voice command.