AI Device Multi-Mode Operation via Dual Wakeup Words
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition systems in AI devices cannot provide continuous interaction with users as they disable command recognition after processing a single command, leading to misrecognition due to everyday conversation or background noise, and require users to repeatedly input a wakeup word for each new command.
Innovation Solution
An AI device that recognizes a second wakeup word after a first wakeup word to set an operation mode, allowing it to process multiple commands without re-inputting the wakeup word, enabling modes like conversation, multiple requests, specific device calls, and secret modes that extend the speech waiting period for continuous interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the command recognition function is disabled after processing a single command, then misrecognition due to everyday conversation or background noise is prevented, but users must repeatedly input a wakeup word for each new command and continuous interaction is broken
Solution Approach 1:
The patent applies dynamics by making the command recognition function dynamically controllable through a mode setting unit. The system transitions between different operation modes (first mode with disabled recognition, second mode with enabled recognition) based on user input, allowing the recognition function to be selectively activated rather than statically disabled or always on. This resolves the contradiction by enabling continuous interaction in the second mode while maintaining reliability through selective disabling in the first mode.
Solution Approach 2:
The patent changes the operational parameter of the command recognition function from a fixed state to a variable state controlled by operation modes. By introducing mode parameters (first operation mode vs. second operation mode), the system can adjust the recognition function's sensitivity and state according to different user needs, thereby balancing reliability and ease of operation.
2Ease of operation
If the command recognition function is always enabled, then continuous interaction and multiple commands can be processed without re-inputting wakeup word, but misrecognition occurs due to everyday conversation or background noise
Solution Approach 1:
The system dynamically adjusts the command recognition function's state based on the current operation mode. In the second operation mode, the recognition function remains enabled for continuous interaction, while in the first operation mode, it is disabled to prevent misrecognition. This dynamic control allows the system to adapt to different interaction scenarios.
Solution Approach 2:
The mode setting unit acts as an intermediary that controls when the command recognition function should be active or inactive. By introducing this intermediate control layer, the system mediates between the need for continuous interaction and the need to prevent misrecognition, allowing users to select the appropriate mode based on the situation.
3Reliability
If users must repeatedly input a wakeup word for each command, then misrecognition is prevented, but the time required for multiple commands increases and interaction becomes less natural
Solution Approach 1:
The patent enables continuity of useful action by allowing the command recognition function to remain enabled during the second operation mode. Users can issue multiple commands without repeatedly inputting the wakeup word, maintaining continuous interaction flow. The useful action of command recognition continues uninterrupted while the mode ensures accuracy is maintained through selective operation.
Solution Approach 2:
The system uses periodic mode switching where the command recognition function is periodically enabled or disabled based on the operation mode. In the second mode, the recognition function operates continuously for a period, allowing multiple commands without wakeup word repetition, thereby reducing time loss while maintaining reliability through the periodic nature of mode-based control.
Data Source
AI summary
An artificial intelligence device is provided. The artificial intelligence device according to an embodiment of the present disclosure includes: an input unit configured to receive a speech input; and a processor configured to operate in an interaction mode if a second wakeup word for setting an operation mode is recognized after a first wakeup word for calling the artificial intelligence device is recognized, and process one or more commands received after the second wakeup word according to the operation mode indicated by the second wakeup word.