Microphone-Based Electronic Device for Multi-User Sleep-State Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to distinguish and analyze the sleeping states of multiple users accurately when they are sleeping together, as conventional methods fail to differentiate between their breathing sounds effectively.
Innovation Solution
An electronic device equipped with a microphone, memory, and processors that utilize neural network models to identify and analyze breathing sounds, employing embedding vectors and latent space analysis to distinguish users and determine their sleeping states, and control external devices based on the analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional breathing sound analysis methods are used, then the system is simple and does not require additional sensors, but it cannot distinguish breathing sounds of multiple users sleeping together
Solution Approach 1:
The patent transforms breathing sound analysis from time-domain to frequency-domain by applying Fast Fourier Transform (FFT), converting raw audio signals into spectral representations. This parameter transformation enables the system to distinguish multiple users' breathing sounds by analyzing frequency characteristics, resolution, and spectral patterns that are not apparent in the time domain, thereby improving measurement precision without adding physical sensors.
Solution Approach 2:
The patent introduces a spectral dimension by converting one-dimensional time-series breathing sound signals into two-dimensional spectrograms (frequency vs. time). This dimensional transformation allows the system to capture and analyze frequency resolution information, enabling differentiation of multiple users' breathing patterns based on their unique spectral signatures, thus resolving the limitation of conventional single-dimensional analysis.
2Measurement precision
If additional sensors are added to improve measurement accuracy, then breathing sound distinction improves, but device cost and complexity increase
Solution Approach 1:
The patent makes the existing microphone serve multiple functions: it not only captures audio for general purposes but also performs spectral analysis for breathing sound differentiation. By applying signal processing techniques (FFT, spectrogram generation) to the microphone's output, the system enables the single sensor to perform both general audio detection and specific breathing analysis, eliminating the need for additional specialized sensors while maintaining measurement precision.
Solution Approach 2:
The patent replaces the need for additional physical sensors with computational methods. Instead of using multiple microphones or specialized acoustic sensors to differentiate users, the system uses digital signal processing (FFT, spectral analysis, machine learning algorithms) to extract distinguishing features from the single microphone's output, substituting mechanical/sensor-based differentiation with algorithm-based differentiation.
3Measurement precision
If spectral analysis and machine learning are applied to distinguish multiple users, then user identification accuracy improves, but processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary processing by generating spectrograms and extracting spectral features before applying machine learning classification. By pre-computing the frequency-domain representation and identifying key spectral characteristics (such as frequency resolution patterns and spectral centroids) beforehand, the system reduces the computational burden during real-time user identification, allowing faster processing while maintaining high identification accuracy through the use of pre-extracted meaningful features.
Data Source
AI summary
An electronic device is provided. The electronic device includes a microphone, memory storing one or more computer programs, and one or more processors communicatively coupled to the microphone, and the memory, wherein the one or more computer programs include computer-executable instructions that, when executed by the one or more processors individually or collectively, cause the electronic device to store registration information on breathing sounds of a plurality of users in the memory, based on receiving an audio signal through the microphone, obtain information on a breathing sound of a user based on the audio signal, compare the information on the breathing sound with the registration information, identify at least one user corresponding to the information on the breathing sound among the plurality of users, and based on identifying the at least one user, obtain an analysis result for the sleeping states of each of the at least one user based on information corresponding to each of the at least one user in the information on the breathing sound.