Dereverberation Gain Control for Voice Communication Audio Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice communication systems face challenges in improving audio quality due to reverberation, which affects the intelligibility of speech and is not effectively addressed by current dereverberation algorithms that can impact other voice processing components like noise reduction.
Innovation Solution
A system that performs dereverberation using a two-channel algorithm, calculating coherence-to-diffuse ratio (CDR) values and deriving dereverberation gain values to reduce reverberation and diffuse noise, while adjusting noise reduction processing based on signal-to-noise ratio (SNR) conditions, ensuring minimal impact on other voice processing components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dereverberation algorithms are applied to reduce reverberation, then audio quality and speech intelligibility are improved, but noise reduction performance deteriorates
Solution Approach 1:
The system dynamically adjusts the dereverberation gain based on real-time analysis of the audio signal characteristics, specifically monitoring the speech activity detection (VAD) states and signal energy levels. This allows the dereverberation processing to be applied adaptively - stronger when speech is present and reverberation is problematic, reduced or disabled when noise reduction is the priority, thus resolving the contradiction between improving speech intelligibility and maintaining noise reduction performance
Solution Approach 2:
The system changes the dereverberation parameter (gain value) based on the detected audio conditions. By analyzing the spectral characteristics and energy distribution of the input signal, the system adjusts the dereverberation strength to optimize speech intelligibility in reverberant environments while preventing degradation of noise reduction performance in noisy environments
2Reliability
If aggressive noise reduction processing is applied, then noise levels are reduced, but speech quality and naturalness deteriorate
Solution Approach 1:
The noise reduction processing intensity is dynamically controlled based on the dereverberation gain and speech activity detection results. When dereverberation is actively reducing reverberation (indicating a reverberant environment), the noise reduction gain is adjusted to be less aggressive, preserving speech quality. When dereverberation is minimal (indicating a less reverberant environment), noise reduction can be applied more aggressively, thus dynamically balancing noise reduction performance with speech quality preservation
Data Source
AI summary
A system configured to improve audio processing by performing dereverberation and noise reduction during a communication session. The system may apply a two-channel dereverberation algorithm by calculating coherence-to-diffuse ratio (CDR) values and calculating dereverberation (DER) gain values based on the CDR values. While the device calculates the DER gain values prior to performing acoustic echo cancellation (AEC) processing, the device applies the DER gain values after performing residual echo suppression (RES) processing in order to avoid excessive attenuation of the local speech. To improve output speech quality, the device does not apply the DER gain values for nonreverberant signals, when a signal-to-noise ratio (SNR) value is too low, and/or when far-end talk (e.g., remote speech) is present. Dereverberation processing is further improved by using frequency dependent parameters to calculate the DER gain values and by adjusting other gain values when the DER gain values are applied.