Electronic Device Voice AI During Calls via ASR and NLU
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices do not support the activation of intelligent systems during a call, preventing users from obtaining or sharing information with another party using AI services while on the phone.
Innovation Solution
The electronic device is equipped with a communication circuit, microphone, speaker, and processor that enable voice-based intelligent services during a call, performing automatic speech recognition and natural language understanding to identify and transmit user inputs to an external server for AI processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the electronic device performs a call with another electronic device, then communication functionality is maintained, but the device cannot activate intelligent services during the call
Solution Approach 1:
The system dynamically adjusts its operational state by detecting call status through the communication circuit and conditionally activating the intelligent service based on whether a call is currently active. The processor transitions between different operational modes (call-only mode vs. call-plus-intelligent-service mode) based on real-time conditions, enabling versatile functionality without permanent system complexity.
Solution Approach 2:
The processor acts as an intermediary that coordinates between the communication circuit (handling calls) and the intelligent service module (handling voice processing). It manages the interaction by receiving voice inputs during calls, determining whether to process them through the intelligent service, and coordinating the flow of information between these two functional blocks without causing system conflict.
2Productivity
If the device processes voice input during a call through ASR and NLU, then intelligent information processing is enabled, but additional processing resources and time are required
Solution Approach 1:
The system performs preliminary actions by continuously monitoring the call status through the communication circuit before attempting to activate intelligent services. The processor prepares the intelligent service module in advance by checking system resources and call state, so when voice input is received, the service can be activated immediately without delay for resource allocation or state verification.
Solution Approach 2:
The patent replaces traditional mechanical voice processing (simple voice commands) with advanced AI-based processing (ASR and NLU). The automatic speech recognition converts spoken language to text, and natural language understanding interprets the user's intent, enabling more sophisticated information processing during calls compared to basic voice recognition systems.
3Adaptability or versatility
If the electronic device activates intelligent service during a call, then multiple functions are supported simultaneously, but system resource management becomes more complex
Solution Approach 1:
The electronic device achieves multi-functionality by enabling the intelligent service to operate concurrently with the call function. The processor is configured to handle both communication tasks (through the communication circuit) and intelligent service tasks (through the microphone and voice processing pipeline) simultaneously, making a single device capable of multiple functions without requiring separate dedicated systems.
Solution Approach 2:
The system implements self-service by automatically detecting when a call is active and conditionally enabling or disabling the intelligent service based on system state. The processor monitors resource availability and call status, and autonomously manages the activation of ASR and NLU functions without requiring manual user configuration, thereby simplifying resource management despite multiple functions being supported.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
According to various embodiments of the disclosure, an electronic device and a method of executing a function of the electronic device comprise at least one communication circuit, a speaker, a microphone, at least one processor operatively connected with the communication circuit, the speaker, and the microphone, and a memory operatively connected with the at least one processor, wherein the memory stores instructions which, when executed by the at least one processor cause the electronic device to: perform a call with a first external electronic device through the communication circuit, activate a voice-based intelligent service while the call is performed, identify a voice input through the microphone while the call is performed, and transmit data associated with the voice input to an external server through the communication circuit at least in part for automatic speech recognition (ASR) and/or natural language understanding (NLU).