Digital AGC for Video Conferencing Signal Stability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional digital automatic gain control (AGC) methods in video conference systems suffer from instability in sound transmission due to varying speech signal amplitudes, leading to distortion and increased noise, particularly when using envelope-based gain calculations and voiceprint recognition for user-specific settings.

Innovation Solution

A digital AGC method that calculates a speech probability and performs speech envelope estimation to adjust signal gains dynamically, using a first gain for amplitude stabilization and a second gain to reduce noise, with optional dynamic compression to prevent voice cracking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Stability of the object's composition

If envelope-based gain calculation is used to adjust signal amplitude, then signal amplitude stabilization is improved, but signal distortion increases due to rapid speech signal amplitude changes

Engineering Contradiction:
Improvesignal amplitude stabilityVSAvoidsignal fidelity
Core Design Contradiction:
Stability of the object's compositionVSManufacturing precision

Solution Approach 1:

The patent implements dynamic gain adjustment by continuously tracking speech signal characteristics and adapting the gain value in real-time. The system transitions from static envelope-based gain to dynamic gain control that responds to speech activity detection, thereby stabilizing amplitude while preserving signal fidelity through adaptive processing.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter used for gain calculation from simple signal envelope to speech activity-based parameters. By detecting speech activity and adjusting gain according to speech presence rather than raw envelope amplitude, the system achieves both amplitude stabilization and reduced distortion.

Inventive Principle:
Principle #35Parameter changes

2Stability of the object's composition

If large gain is applied to non-speech signals with small envelope values, then signal amplitude is boosted, but output noise increases significantly

Engineering Contradiction:
Improvesignal amplitude consistencyVSAvoidoutput noise
Core Design Contradiction:
Stability of the object's compositionVSObject-generated harmful factors

Solution Approach 1:

The patent extracts speech activity information from the signal and uses it to control gain application. By separating speech detection from amplitude control, the system applies gain selectively based on speech presence rather than uniformly to all signals, thereby avoiding noise amplification during non-speech periods while maintaining amplitude consistency during speech.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements feedback control by continuously monitoring speech activity and adjusting gain accordingly. The speech activity detector provides feedback that modulates the gain value, creating a closed-loop system that increases gain during speech and reduces it during non-speech, thus achieving amplitude consistency without excessive noise amplification.

Inventive Principle:
Principle #23Feedback

3Reliability

If voiceprint recognition is used for user identification and historical gain retrieval, then user-specific call quality is improved, but calculation complexity and processing time increase greatly

Engineering Contradiction:
Improvecall quality consistencyVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent performs preliminary speech activity detection and gain calculation based on current signal characteristics rather than relying on complex voiceprint recognition for every processing decision. By pre-processing signals with simpler speech detection algorithms and only using historical data when speech is detected, the system achieves consistent call quality with reduced computational complexity.

Inventive Principle:
Principle #10Preliminary action

4Stability of the object's composition

If conventional AGC algorithms are used to stabilize signal amplitude, then transmit volume control is achieved, but voice cracking occurs during rapid amplitude transitions

Engineering Contradiction:
Improvetransmit volume stabilityVSAvoidvoice signal quality
Core Design Contradiction:
Stability of the object's compositionVSManufacturing precision

Solution Approach 1:

The patent implements dynamic gain control that adapts to speech characteristics in real-time. By making the gain parameter dynamic and speech-dependent rather than static, the system stabilizes transmit volume while avoiding voice cracking through adaptive response to actual speech conditions.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the gain parameter based on speech activity detection results. By modifying the gain value according to detected speech characteristics rather than using fixed conventional AGC parameters, the system achieves volume stability while preserving voice signal quality and preventing distortion.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3792918B1Digital automatic gain control method and apparatus
Publication Date: 2023.11.01 YEALINK (XIAMEN) NETWORK TECHNOLOGY CO LTD
  • EP3792918B1 patent drawingFigure 1
  • EP3792918B1 patent drawingFigure 2
  • EP3792918B1 patent drawingFigure 3

AI summary

A digital automatic gain control (AGC) method and apparatus are disclosed in this application. A specific implementation of the method includes: calculating a speech probability Pn of each segment of signal data in a piece of read signal data by using a speech probability calculation model; performing a speech envelope estimation on the signal data to obtain a speech envelope amplitude; calculating a first gain according to a deviation between the speech envelope amplitude and an anticipated amplitude; classifying the signal data based on the speech probability Pn, and counting the number of noise occurrences in the signal data based on a classification result, to calculate a second gain; and performing signal gain amplitude adjustment on the signal data by using the first gain and the second gain. This implementation helps achieve an overall amplification effect and reduce distortion caused by automatic gain processing. Moreover, in the case of user change, the signal gain amplitude can also be adjusted quickly in response to different users. Meanwhile, in the case of excessive ambient noise, the apparatus can also adjust the gain automatically, to reduce amplification of the noise.