Non-Speech Audio Signal Recognition in Electronic Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices struggle to accurately recognize and utilize non-speech audio signals, which are difficult to interpret without specific meaning assignment, limiting their functionality compared to speech audio signals.
Innovation Solution
An electronic device equipped with microphones, processors, and memory that performs speech and non-speech audio processing to identify non-speech audio signal patterns, allowing users to register meanings and commands for non-speech audio signals, enabling precise recognition and control through separate processing methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If speech audio processing is used, then speech recognition accuracy is improved, but non-speech audio signal recognition capability deteriorates
Solution Approach 1:
The patent divides audio signal processing into separate processing paths: speech audio processing for speech signals and non-speech audio processing for non-speech signals. This segmentation allows each processing path to be optimized independently, resolving the contradiction between speech recognition accuracy and non-speech recognition capability
Solution Approach 2:
The patent introduces an audio signal type determination unit as an intermediary that identifies whether the input signal is speech or non-speech before routing to appropriate processing. This mediator enables the system to apply different processing strategies based on signal type, improving both speech and non-speech recognition
2Measurement precision
If separate non-speech audio processing is implemented, then non-speech audio signal recognition rate is improved, but device complexity increases
Solution Approach 1:
The patent implements a unified audio processing system that handles both speech and non-speech signals through a single architecture with branching processing paths. The audio signal type determination unit enables one system to perform multiple functions (speech processing and non-speech processing), reducing overall device complexity while improving non-speech recognition
Solution Approach 2:
The patent changes the processing parameters and algorithms based on the identified audio signal type. When non-speech signals are detected, the system switches to non-speech audio processing with appropriate parameters, achieving high recognition rates without requiring permanently complex processing paths for all signals
Data Source
AI summary
An electronic device is provided. The electronic device comprises a speaker, a plurality of microphones, at least one processor operatively connected with the speaker and the plurality of microphones, and a memory operatively connected with the at least one processor, wherein the memory is configured to store instructions which, when executed, cause the at least one processor to perform speech audio processing or non-speech audio processing on audio signals received via the plurality of microphones, upon obtaining a non-speech audio signal based on the speech audio processing or the non-speech audio processing, identify a non-speech audio signal pattern corresponding to the non-speech audio signal, obtain a non-speech audio signal-based first command based on the identified non-speech audio signal pattern, and perform at least one action corresponding to the obtained non-speech audio signal-based first command.