Two-Stage Digital AGC for Stable Conferencing Audio Levels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video conferencing systems face challenges in maintaining a stable and consistent audio signal level, leading to unpredictable audio quality due to varying speech levels, which can result in unpleasant auditory experiences for recipients.
Innovation Solution
A two-stage automatic gain control (AGC) module is implemented, comprising a long-term level estimator and a short-term level estimator, which work together to gradually and instantaneously adjust the audio signal to a target level, ensuring consistent audio quality by applying appropriate digital gains based on estimated signal levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If a single-stage AGC is used to adjust audio signal level, then the device complexity is low, but the audio signal level stability is poor
Solution Approach 1:
The AGC module is divided into two distinct stages: a long-term level estimator that provides gradual gain adjustment for overall stability, and a short-term level estimator that provides instantaneous gain adjustment for rapid response. This segmentation allows each stage to specialize in different temporal characteristics, resolving the contradiction between stability and complexity by distributing functions across multiple components rather than using a single complex stage.
2Speed
If gain is adjusted instantaneously to reach target level, then the response speed is fast, but noise and interference increase
Solution Approach 1:
The long-term level estimator performs preliminary gain adjustment gradually over time to reach the target level, preparing the signal in advance. The short-term level estimator then makes fine instantaneous adjustments only when necessary, based on recent signal characteristics. This preliminary action approach allows fast response when needed while avoiding excessive noise by using gradual adjustment as the base.
Solution Approach 2:
The system periodically updates the long-term level estimate over extended time intervals and combines it with short-term periodic updates. This periodic action at different time scales allows the system to maintain stability through long-term averaging while responding to immediate changes, reducing noise by not continuously adjusting at maximum gain.
3Object-affected harmful factors
If gain is adjusted gradually to reach target level, then noise and interference are reduced, but the response speed is slow
Solution Approach 1:
The two-stage architecture segments the gain adjustment function into long-term gradual adjustment (reducing noise) and short-term instantaneous adjustment (improving speed). The long-term stage handles overall level stabilization with noise reduction, while the short-term stage provides rapid response to sudden signal changes, thus resolving the contradiction between gradual adjustment and response speed.
4Illumination intensity
If high gain is applied to amplify weak speech signals, then the speech level is sufficient, but overamplification occurs
Solution Approach 1:
The system continuously estimates signal levels using both long-term and short-term estimators, creating a feedback mechanism that monitors the current signal state. Based on this feedback, the gain is dynamically adjusted: the long-term estimator provides gradual gain increase for weak signals, while the short-term estimator prevents overamplification by detecting rapid level changes and adjusting gain accordingly, ensuring speech signals reach sufficient levels without distortion.
Data Source
AI summary
Disclosed are systems and methods for automatic control of gain in audio and video conferencing applications to maintain a predetermined and stable audio level. In one embodiment, a first stage applies a first stage gain based on a long-term estimate of signal level, while a second stage gain, based on a short-term estimate of signal level assists the first stage gain to achieve a target level. Some embodiments of long-term level estimation utilize statistical analysis of a buffer to validate or arrive at a more accurate long-term signal level estimate.