Recursive Intent Parsing for Compound Voice Commands
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing natural language processing systems struggle to accurately interpret user commands that are compound or complex sentences, often failing to execute all intended functions when multiple actions are conveyed in a single sentence.
Innovation Solution
A natural language processing apparatus and method that determines whether a user's speech command is a compound or complex sentence, recursively calling the natural language understanding module to process each part of the sentence, using a control module to identify intents and slots, and generating appropriate control signals based on consistency checks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing natural language processing apparatus extracts a single intention from a single input sentence, then the processing is simple and fast, but multiple desired functions may not be performed when user inputs a speech command for a plurality of functions as a single sentence
Solution Approach 1:
The patent segments a compound or complex sentence into multiple sub-sentences or clauses, each representing a distinct function or intention. The control module identifies compound/complex sentences and divides them into separate processing units that can be individually analyzed by the natural language understanding module, enabling multiple functions to be extracted and executed from a single user input.
Solution Approach 2:
The patent implements a nested processing structure where the natural language understanding module is recursively called within the control module. When a compound or complex sentence is detected, the system nests another instance of the intent extraction process within the original processing flow, allowing hierarchical analysis of multiple intentions embedded within a single sentence structure.
2Reliability
If the natural language understanding module is recursively called for compound or complex sentences, then all plurality of functions are executed, but the processing time and computational resources increase
Solution Approach 1:
The control module performs preliminary analysis to identify whether an input sentence is compound or complex before initiating full recursive processing. By detecting sentence structure characteristics in advance, the system prepares appropriate processing strategies and avoids unnecessary recursive calls for simple sentences, thereby reducing overall processing time while maintaining accurate recognition for complex cases.
Solution Approach 2:
The natural language understanding module is designed to be self-contained and reusable through recursive calls. Each invocation of the module independently processes a segment of the input, extracting intentions and slots without requiring external assistance. This self-service capability allows the same module to handle multiple levels of analysis efficiently, reducing the need for additional specialized components.
3Extent of automation
If matching probability thresholds are used to determine compound or complex sentences, then the determination is automated and consistent, but false positives or negatives may occur
Solution Approach 1:
The control module uses matching probability thresholds as feedback mechanisms to determine whether a sentence is compound or complex. By comparing the calculated matching probabilities against predefined thresholds, the system automatically adjusts its classification decisions. This feedback loop enables consistent automation while allowing for threshold optimization to minimize false positives and negatives based on actual performance data.
Data Source
AI summary
The present disclosure provides a natural language processing apparatus and a natural language processing method that may determine whether a user's speech command is a compound sentence or a complex sentence based on an output of a natural language understanding module, and when the user's speech command is a compound sentence or a complex sentence, recursively call the natural language understanding module, thereby executing all of the plurality of functions expressed as a single sentence.