Acoustic Signal Processing Apparatus for Noise Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional teleconference systems face challenges in isolating human voices from background noise, as directional microphones often pick up unwanted sounds like paper flipping and chair movement, degrading audio communication quality.
Innovation Solution
An acoustic signal processing apparatus that uses omnidirectional microphones to collect sound from multiple directions, generates effective signals by beamforming, calculates features based on these signals, and selects the target direction to enhance voice capture while reducing noise, thereby improving audio communication precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If directional microphones are disposed near conference participants and turned ON for speaking persons, then human voices can be collected with higher precision, but noise sounds such as paper flipping and chair movement are also collected and transmitted
Solution Approach 1:
The patent segments the acoustic space into multiple examination directions and processes signals from each direction separately. By dividing the 360-degree space into discrete sectors and analyzing each independently, the system can identify and select only the target direction containing human voice, while excluding other directions containing noise sources like paper flipping and chair movement.
Solution Approach 2:
The patent implements dynamic beamforming by continuously adjusting the beam direction based on real-time feature analysis of acoustic signals. The system dynamically switches between different examination directions and updates the beam pattern adaptively, allowing it to track moving speakers while maintaining focus on the current target direction and rejecting noise from other directions.
2Area of stationary object
If microphones are disposed at a plurality of positions and output signals are mixed, then sound collection coverage is improved, but noise sounds from multiple directions are also collected and transmitted
Solution Approach 1:
The patent segments the multi-directional acoustic signals into separate examination directions before processing. Instead of mixing all microphone signals together, the system divides them into spatial sectors, processes each sector independently to identify human voice characteristics, and then selects only the target direction for output, thereby maintaining wide coverage while rejecting noise from non-target directions.
Solution Approach 2:
The patent processes acoustic signals for all examination directions (excessive action) but selectively outputs only the target direction containing human voice (partial action). By performing feature analysis and selection on multiple directions and then outputting only the relevant portion, the system achieves comprehensive noise rejection while maintaining complete voice capture capability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively isolates human voices from background noise, enhancing audio communication quality by selectively focusing on the target sound source and reducing unwanted noise transmission, thus providing a more comfortable and precise communication experience.
Implementation Method 1
a plurality of sound receivers receives sound from a plurality of examination directions in a space and outputs acoustic signals of a plurality of channels
Implementation Method 2
generates an effective signal corresponding to sound coming from each one of the examination directions based on the acoustic signals of the plurality of channels
Data Source
AI summary
An acoustic signal processing apparatus includes circuitry to generate, when a plurality of sound receivers receive sound from a plurality of examination directions in a space and outputs acoustic signals of a plurality of channels, an effective signal corresponding to sound coming from each one of the examination directions based on the acoustic signals of the plurality of channels for each one of the examination directions, calculate a feature for each one of the examination directions based on the effective signal generated for each one of the examination directions, and select a target direction from the plurality of examination directions in the space based on the feature calculated for each one of the examination directions.