Speech Recognition Server for Home Appliances
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech recognition systems for home appliances require multiple speech commands to set complex operation conditions, making it inconvenient for users to control advanced features, leading to user frustration and abandonment of functionality.
Innovation Solution
A speech recognition system that interprets a single sentence with multiple intents to collectively set multiple setting items, using a server to generate instruction formulas, derivative sentences, and compare them with registered speech command data to control home appliances efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If multiple speech commands are used to set each setting item individually, then each setting can be controlled precisely, but the operation becomes complicated and time-consuming
Solution Approach 1:
The patent combines multiple setting items into a single speech command input. Instead of requiring separate speech commands for each setting item (e.g., number of rinsing operations, water level), the system allows users to input all settings in one natural language sentence, which is then parsed to extract multiple intents and set all parameters simultaneously.
Solution Approach 2:
The patent segments a single speech command into multiple intents by analyzing the natural language input. The speech recognition system divides the unified speech input into distinct setting items (e.g., separating 'number of rinsing operations' from 'water level' from 'washing course'), allowing each setting to be processed and configured independently while originating from a single user input.
2Ease of operation
If multiple speech commands are required for complex functions, then each function can be set accurately, but the operation complexity increases making it difficult to understand without user guidebooks
Solution Approach 1:
The patent implements a universal speech recognition framework that can handle multiple setting items and various home appliance functions through a single natural language interface. The system is designed to interpret diverse speech inputs and map them to multiple different functions and parameters, making the control method consistent across different appliance types and functions.
Solution Approach 2:
The patent replaces the mechanical key manipulation system with a speech-based natural language processing system. Instead of requiring users to press multiple keys in specific sequences on a control panel, the system substitutes this mechanical interaction with speech recognition and natural language interpretation, making the operation more intuitive and easier to understand.
3Reliability
If traditional key manipulation is used for setting, then precise control is achieved, but user frustration increases leading to abandonment of functions
Solution Approach 1:
The patent replaces the mechanical key manipulation system with a speech-based natural language processing system. Instead of requiring users to press multiple keys in specific sequences on a control panel, the system substitutes this mechanical interaction with speech recognition and natural language interpretation, making the operation more intuitive and easier to understand.
Solution Approach 2:
The system automatically parses and interprets the natural language speech input to extract multiple settings and intents without requiring users to manually navigate through complex menus or press multiple keys. The speech recognition system self-services by automatically understanding the user's intent and configuring multiple parameters based on the single speech input.
Data Source
AI summary
Provided is a system, server, and method for speech recognition capable of collectively setting a plurality of setting items for device control through an utterance of a single sentence provided in the form of natural language. The system includes: a home appliance configured to receive a speech command that is generated through an utterance of a single sentence for control of the home appliance; and a server configured to receive the speech command in the single sentence from the home appliance and interpret the speech command of the single sentence through multiple intent determination.