Adaptive Gain Control with Speech-Aware Output Compensation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automatic gain control systems operate in isolation, potentially suppressing desired signal content in input signals, leading to inaccurate results in speech processing and wireless voice networks due to lack of consideration for input signal content.

Innovation Solution

An automatic gain control system that determines input signal content and applies compensating gains to maintain consistent desired signal levels, attenuating input signals when necessary and compensating output signals based on the presence of desired or undesired signal components.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If automatic gain control attenuates input signal to keep level within range, then received signal level is maintained, but desired speech signal content is suppressed

Engineering Contradiction:

Improvesignal level controlVSAvoidspeech signal content

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The system segments the input signal into desired speech components and undesired noise components. By identifying and separately processing these components, the system can attenuate undesired content while preserving desired speech, thus resolving the contradiction between maintaining signal level and preserving speech content.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system uses feedback mechanisms to monitor the input signal characteristics and dynamically adjust gain control parameters. By continuously analyzing signal content and adjusting attenuation accordingly, the system maintains signal levels within range while minimizing suppression of desired speech content.

Inventive Principle:

Principle #23Feedback

2Device complexity

If automatic gain control operates in isolation without knowledge of signal content, then gain control is simple, but downstream processing accuracy deteriorates

Engineering Contradiction:

Improvegain control systemVSAvoidspeech recognition accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system performs preliminary analysis of the input signal to identify desired speech content and undesired noise before applying gain control. This preliminary action enables the system to make informed gain adjustments that preserve speech recognition accuracy while maintaining simple overall system architecture.

Inventive Principle:

Principle #10Preliminary action

3Manufacturing precision

If input signal level exceeds processing bound, then signal distortion occurs, but desired signal content may be attenuated incorrectly

Engineering Contradiction:

Improvesignal processing qualityVSAvoiddesired signal content

Core Design Contradiction:

Manufacturing precisionVSLoss of information

Solution Approach 1:

The system applies different processing qualities to different parts of the signal. By identifying regions with desired speech content versus undesired noise, the system applies localized gain control that maintains processing quality for speech portions while allowing different treatment for noise portions, preventing incorrect attenuation of desired content.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS8116485B2Adaptive gain control system

Publication Date: 2012.02.14 BLACKBERRY LTD

AI summary

An automatic gain control system maintains desired signal content level, such as voice, in an output signal. The system includes automatic gain control over an input signal, and compensates the output signal based on input signal content. When the input signal level exceeds an upper or lower processing threshold level, or is distorted (e.g., clipped), the system applies a gain to the input signal level. The system may compensate for the gain in the output signal when the input signal includes desired signal content.