NLU Voice Recognition Handling Obsequious Expressions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice recognition systems face inaccuracies with unclear or soft-spoken inputs, particularly in children, and require additional training and complex architecture to handle polite expressions, leading to reduced accuracy and user frustration.
Innovation Solution
A natural language model-based voice recognition system that minimally trains and adapts for legacy systems, using a binary classifier to distinguish between queries with and without obsequious expressions, allowing for plug-and-play modularity and improved recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If voice recognition systems require polite expressions (please, thank you) to be included in commands, then device responsiveness to polite speech improves, but speech recognition accuracy deteriorates due to reduced match scores
Solution Approach 1:
The system segments the command processing into two independent parts: a polite expression detection module that identifies polite speech, and a traditional speech recognition module that processes the core command. This segmentation allows the system to handle polite expressions without interfering with the accuracy of command recognition, resolving the contradiction between responsiveness to polite speech and speech recognition accuracy.
2Adaptability or versatility
If the system is trained to recognize queries with obsequious expressions, then ability to handle polite speech improves, but system architecture complexity increases
Solution Approach 1:
The system introduces an intermediary polite expression detection module that acts as a mediator between the user's polite speech and the traditional speech recognition system. This intermediary component identifies and handles polite expressions separately, allowing the core speech recognition architecture to remain unchanged and simple, thus achieving adaptability without increasing overall system complexity.
3Adaptability or versatility
If extensive retraining is performed to accommodate polite expressions, then model understanding of expressions improves, but training time and computational resources increase
Solution Approach 1:
The system performs preliminary action by pre-processing the input speech to detect and separate polite expressions before the main speech recognition process. By identifying polite expressions in advance and handling them separately, the system avoids the need for extensive retraining of the core speech recognition model, thereby reducing training time and computational resources while maintaining expression understanding capability.
Data Source
AI summary
Systems and methods for training a classifier binary model of a natural language understanding (NLU) system are disclosed herein. A determination is made as to whether a text string, with a content entity, includes an obsequious expression. In response to determining the text string includes an obsequious expression, a determination is made as to whether the obsequious expression describes the content entity. The model is trained based on a determination of at least one of: an absence of an obsequious expression in response to determining the obsequious expression describes the content entity; a presence of an obsequious expression in response to determining the obsequious expression describes the content entity; an absence of an obsequious expression in response to determining the obsequious expression does not describe the content entity, and a presence of an obsequious expression in response to determining the obsequious expression does not describe the content entity.