Adaptive Voice Signal Separation Using DOA Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices face challenges in clearly distinguishing voice signals from noise, leading to incomplete removal of noise or inclusion of noise in voice signals, especially when the environment changes, resulting in unsatisfactory audio data for users.
Innovation Solution
The device employs multiple microphones and a processor to estimate the probability of voice signal existence, obtain correlation information, and use direction of arrival (DOA) estimation to separate noise from voice signals, effectively removing noise while minimizing signal loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the electronic device uses a single microphone or fixed directional filtering, then the device structure is simple, but the device cannot adapt to environmental changes and fails to clearly distinguish voice signals from noise
Solution Approach 1:
The patent implements dynamic adaptability by continuously estimating the direction of arrival (DOA) of sound signals and updating beamforming weights in real-time based on environmental changes. The system adjusts its noise cancellation behavior dynamically rather than using fixed filtering, allowing it to adapt to changing acoustic environments while maintaining a relatively simple hardware structure.
Solution Approach 2:
The patent replaces complex mechanical or hardware-based directional microphones with software-based signal processing techniques including DOA estimation algorithms and adaptive beamforming. This substitution allows the device to achieve sophisticated noise cancellation and voice signal separation using computational methods rather than complex physical structures.
2Reliability
If the electronic device removes all signals except in a fixed direction, then noise removal is simple, but voice signals may be removed together with noise when direction changes
Solution Approach 1:
The patent employs feedback mechanisms by continuously monitoring the acoustic environment, estimating DOA of sound sources, and using this information to adjust beamforming weights in real-time. This closed-loop system ensures that voice signals are preserved while noise is removed, with the processing complexity justified by the significant improvement in voice signal reliability.
Solution Approach 2:
The patent changes the parameters of signal processing by dynamically adjusting beamforming weights based on estimated DOA and signal characteristics. Rather than using fixed directional filtering, the system adapts its frequency response and spatial filtering parameters in real-time to maintain voice signal integrity across varying environmental conditions.
3Measurement precision
If the electronic device uses complex noise removal algorithms, then noise removal effectiveness is improved, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary DOA estimation and beamforming weight calculation before actual noise removal processing. By pre-computing the spatial filtering parameters based on current environmental conditions, the system reduces the computational burden during real-time noise removal, achieving high noise removal effectiveness without excessive processing delays.
Data Source
AI summary
According to various embodiments, an electronic device includes a plurality of microphones, and a processor electrically connected to the plurality of microphones, wherein the processor may obtain audio signals through the plurality of microphones, estimate a probability of existence of a voice signal included in the obtained audio signals, obtain correlation information between the audio signals based on the probability of existence of the voice signal and/or the obtained audio signals, obtain voice blocking information based on the correlation information or a direction of arrival (DOA) estimation, obtain a first signal among the audio signals based on the audio signals, the correlation information, and the voice blocking information, obtain a second signal including the voice signal among the audio signals, and obtain a noise-removed voice signal by removing the first signal from the second signal. In addition, it is possible to implement various embodiments understood through the disclosure.