Voice-Activated Device Connection via Speaker Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices face challenges in selecting the appropriate external device for voice commands, particularly in multi-user environments, leading to potential privacy violations when the wrong device is activated for making calls, especially in scenarios like multiple family members using an AI speaker.
Innovation Solution
Implementing personalized device connection through speaker recognition, where the electronic device identifies and connects to the correct external device based on the speaker's voice, ensuring accurate and secure voice command execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the electronic device connects to multiple external electronic devices, then the device can serve multiple users and functions, but it becomes difficult to select the appropriate external device for voice commands, leading to privacy violations
Solution Approach 1:
The patent segments the multi-user environment by creating distinct user profiles with unique voiceprints. Each user's voice characteristics are separately stored and recognized, allowing the system to segment and identify which user is speaking, thereby selecting the appropriate external device for that specific user without confusion with other users' devices
Solution Approach 2:
The patent replaces manual device selection (mechanical/user interaction) with automatic voice-based identification. Instead of requiring users to manually select devices or hope for correct random selection, the system uses speaker recognition technology to automatically identify the user and select the corresponding external device, eliminating the selection error problem
2Ease of operation
If the electronic device uses simple device scan for connection, then the connection process is fast and simple, but the wrong external device may be selected in multi-user environments
Solution Approach 1:
The patent performs preliminary actions by registering and storing voiceprint information for each user before actual voice command processing occurs. During the connection setup phase, the system pre-collects and stores unique voice characteristics for each user profile, so that when a voice command is later received, the system can immediately match it against the pre-stored voiceprints to identify the user and select the correct device without requiring additional manual input
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Disclosed is an electronic device including at least one communication circuit, a sound input circuit, a processor operatively connected to the at least one communication circuit and the sound input circuit, and a memory operatively connected to the processor. The electronic device may obtain voice data corresponding to a detected utterance when the utterance is detected using the sound input circuit, may identify speaker information of the voice data based at least on speech recognition of the voice data, may communicatively connect the electronic device to a first external electronic device, using address information of the first external electronic device associated with the speaker information, and may perform an action corresponding to the voice data together with the first external electronic device by using the at least one communication circuit.