Acoustic Echo Cancellation Control via Energy Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio devices face challenges in accurately isolating near-end speech from far-end audio during voice communication, leading to echo distortion and garbled speech processing, especially when near-end speech is present without far-end speech.
Innovation Solution
The system determines the presence of near-end and far-end audio by measuring energy levels and uses this information to toggle the acoustic echo cancellation (AEC) system on and off, ensuring that AEC is only active when far-end audio is present, thereby preventing unwanted cancellation of near-end speech.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If acoustic echo cancellation (AEC) is continuously active to cancel far-end audio, then echo suppression is improved, but near-end speech is distorted or cancelled along with the echo
Solution Approach 1:
The patent applies dynamics by making the AEC system switchable between active and inactive states based on real-time audio conditions. The system dynamically adjusts its operation mode (continuous AEC, interrupted AEC, or AEC bypass) according to whether near-end speech is detected, thereby resolving the contradiction between echo suppression and speech preservation
Solution Approach 2:
The system uses feedback from energy level detectors that continuously monitor both near-end and far-end audio signals. This feedback mechanism triggers state transitions in the AEC system - when near-end speech energy exceeds a threshold, the system receives feedback to interrupt or bypass AEC, preventing speech distortion while maintaining echo cancellation when appropriate
2Object-affected harmful factors
If AEC processing is continuously applied to all audio data, then echo cancellation performance is improved, but system complexity and computational load increase
Solution Approach 1:
The patent implements periodic action by interrupting AEC processing during periods when near-end speech is detected. Instead of continuous processing, the system periodically evaluates audio conditions using energy detectors and switches AEC on or off accordingly, reducing computational load while maintaining effectiveness during far-end audio periods
Solution Approach 2:
The system applies partial action by selectively enabling AEC only when far-end audio is present and near-end speech is absent. This partial application of AEC processing avoids unnecessary computational overhead during periods when echo cancellation is not needed, thereby reducing system complexity and energy consumption
Data Source
AI summary
Techniques for improving acoustic echo cancellation are described. Energy levels of audio data received from a microphone and representing near-end audio and reference audio data representing far-end audio are determined. If near-end audio is detected but far-end audio is not detected, a controller turns of or bypasses an acoustic echo cancellation system until far-end audio is again detected, thereby decreasing or eliminating distortion of the near-end audio by the acoustic echo cancellation system.