Echo Cancellation Circuitry With Comfort Noise Generator
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing echo cancellation techniques face challenges in efficiently removing residual echoes and providing comfort noise, leading to disconcerting quietness and excessive switching between encoding rates, which affects the quality of communication in public telephone networks.
Innovation Solution
The integration of echo cancellation circuitry and variable rate encoding circuitry, where residual echo control signals determine when to replace residual echoes with comfort noise, allowing the speech encoder to adjust encoding rates accordingly, thereby simplifying the encoding process and reducing unwanted noise perception.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the NLP is activated to remove residual echo when the near-end user is not talking, then the echo cancellation effectiveness is improved, but the background noise is also removed resulting in disconcerting quietness
Solution Approach 1:
A comfort noise generator is introduced as an intermediary component that generates artificial comfort noise to replace the removed background noise. This comfort noise serves as a mediator between the NLP echo cancellation function and the far-end user's perception, preventing the disconcerting quietness while maintaining effective echo cancellation.
Solution Approach 2:
The comfort noise generator creates a copy or approximation of the original background noise characteristics. By analyzing the spectral quality and power level of the near-end user's background noise, the system generates comfort noise that closely replicates the original noise profile, making it imperceptible to the far-end user while filling the silence created by NLP.
2Object-affected harmful factors
If the comfort noise generator estimates and generates comfort noise with the same spectral quality and power level as actual background noise, then the perceived noise quality is improved, but the computational complexity and processing overhead increase
Solution Approach 1:
Instead of perfectly replicating all aspects of the original background noise, the comfort noise generator performs partial action by focusing only on the critical spectral quality and power level parameters. This partial replication is sufficient to achieve the desired effect of masking the silence without requiring full-fidelity noise reconstruction, thereby reducing computational complexity.
Solution Approach 2:
The system changes the parameters of the comfort noise generation process by using simplified spectral analysis and power level estimation rather than full-spectrum noise reconstruction. This parameter optimization allows the system to generate adequate comfort noise with reduced computational overhead by focusing on the most perceptually relevant parameters.
3Productivity
If the speech encoder uses variable rate encoding to encode active speech at higher rates and inactive speech at lower rates, then the encoding efficiency is improved, but excessive switching between encoding rates causes unpleasant clicking sounds
Solution Approach 1:
The comfort noise generator provides beforehand cushioning by continuously providing noise during transition periods between active and inactive speech states. This cushioning effect smooths the transitions between different encoding rates, preventing the abrupt changes that cause clicking sounds while maintaining the efficiency benefits of variable rate encoding.
Solution Approach 2:
The comfort noise ensures continuity of useful action by maintaining a continuous noise signal during inactive speech periods and transitions. This continuous noise presence allows the speech encoder to maintain smoother encoding rate transitions without abrupt discontinuities, thereby eliminating clicking sounds while preserving encoding efficiency.
Data Source
AI summary
The present invention provides for echo cancellation circuitry and variable rate encoding circuitry to cooperate with one another to effectively provide comfort noise in an effective and efficient manner. The echo cancellation circuitry will use far-end signals to generate estimated echo signals, which correspond to the actual echo signals appearing in near-end signals. The estimated echo signals are essentially subtracted from the near-end signals in an effort to remove the actual echo signals from the near-end signals. The echo cancellation circuitry will monitor any residual echo signals in the resulting processed near-end signals and provide residual echo control signals that are indicative of whether the residual echo signals should be replaced with comfort noise. The residual echo control signals are used at least in part by the variable rate encoding circuitry to determine the encoding rate to use for encoding different portions of the processed near-end signals.