Voice Device Selection via Multi-Device Signal Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional voice-enabled devices struggle to accurately select the correct IoT device for activation due to differences in microphone position, sensitivity, and recognition models, leading to failure in recognizing the user's speech, especially when devices are not closest to the speech position.
Innovation Solution
A method and apparatus that utilize a pre-trained classification model to analyze voice signals from multiple registered devices, considering recognition situation information such as recognition rate, noise, activation history, and user history, to select the device closest to the speech position for activation, leveraging a 5G network for communication and AI processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the server selects a voice-enabled device based solely on wakeup word recognition results, then the selection process is simple and fast, but the selected device may not be the one the user intends to activate when the closest device fails to recognize the wakeup word
Solution Approach 1:
The system performs preliminary actions by obtaining voice signals from multiple devices before making the final selection. Instead of waiting for recognition results only, the server proactively collects voice signals from devices that may have missed the wakeup word, enabling more accurate distance-based selection while maintaining efficient processing through pre-positioned data collection mechanisms.
Solution Approach 2:
The server acts as an intermediary that mediates between multiple voice-enabled devices and the user's intended target. By collecting and analyzing voice signals from multiple devices, the server intermediates the selection process to determine which device the user most likely intended to activate, resolving the contradiction between simple selection and accurate device identification.
2Adaptability or versatility
If multiple IoT devices transmit voice signals to the server, then the system can select from more options, but the communication overhead and processing time increase
Solution Approach 1:
The system extracts only the necessary voice signals from multiple devices rather than processing all possible data. By selectively obtaining voice signals from devices that are candidates for activation, the system maintains versatile device selection options while minimizing communication overhead and processing time by excluding unnecessary data transmission.
3Reliability
If the system requests voice signals from all registered devices, then the selection accuracy improves, but the communication load and energy consumption increase
Solution Approach 1:
The system applies partial action by requesting voice signals from only the necessary subset of devices rather than all registered devices. By identifying and contacting only those devices that have a reasonable probability of containing the user's intended target, the system achieves sufficient selection accuracy while significantly reducing communication energy consumption compared to querying every device.
Data Source
AI summary
A method and apparatus for selecting a voice-enabled device are disclosed. In the voice-enabled device selecting method, when recognition situation information is obtained from one device, voice signals may be obtained from that device and second devices registered to the same account as that device, and a device that will respond to the wakeup word may be selected based on the voice signals. Thus, even if the closest device the user intends to activate is not able to recognize a wakeup word spoken by the user due to the device's microphone position, that device may be selected as a voice-enabled device. At least one of a voice enable device selecting apparatus, an intelligent computing device, an IoT device, and a server controlling the voice enable device selecting apparatus may be associated with an artificial intelligence (AI) module, an unmanned aerial vehicle (UAV) (or drone), a robot, an augmented reality (AR) device, a virtual reality (VR) device, and a device related to a 5G service.