Adaptive Noise Suppression for Super Wideband Audio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise suppression technologies in wireless communication devices distort music signals by treating them as noise, leading to poor quality music transmission, especially when used in environments with background noise, as they are not capable of handling both speech and music signals effectively up to super wideband bandwidths.
Innovation Solution
Implementing adaptive noise suppression techniques that identify the context of audio data as either speech or music and adjust the noise suppression level accordingly, allowing for minimal distortion of music signals by applying a less aggressive noise suppression when music is detected, enabling proper encoding by Enhanced Voice Services (EVS) codecs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If aggressive noise suppression is applied to improve speech intelligibility, then speech quality is improved, but music signals are distorted and suppressed
Solution Approach 1:
The noise suppression level is made dynamic and adaptive based on the detected audio context. The system switches between different noise suppression levels (first level for speech contexts, second level for music contexts) depending on whether speech or music is detected as the dominant signal type, thereby resolving the contradiction between speech intelligibility and music quality
Solution Approach 2:
The system changes the noise suppression parameter (suppression level) based on the audio context. When speech is detected, aggressive noise suppression is applied; when music is detected, less aggressive noise suppression is applied. This parameter adaptation allows the system to optimize for the current audio type, resolving the quality trade-off
2Reliability
If noise suppression treats all non-speech signals as noise, then speech signals are cleaned up, but music signals are eliminated
Solution Approach 1:
The system applies different noise suppression treatments to different audio contexts locally. Speech contexts receive aggressive noise suppression while music contexts receive minimal or no noise suppression. This localized adaptation based on context classification preserves music signal integrity while maintaining speech clarity
Solution Approach 2:
The system uses audio context detection as feedback to dynamically adjust the noise suppression level. The detected audio context (speech or music) feeds back to control the noise suppression intensity, preventing the loss of music signals while maintaining speech signal clarity
3Reliability
If conventional noise suppression is used in noisy environments, then speech intelligibility is improved, but music transmission quality degrades
Solution Approach 1:
The noise suppression system dynamically adapts its behavior based on the audio context detected in noisy environments. When music is detected as the dominant signal, the system reduces noise suppression intensity to preserve music transmission quality, while still maintaining speech intelligibility when speech is detected
Data Source
AI summary
Techniques are described for performing adaptive noise suppression to improve handling of both speech signals and music signals at least up to super wideband (SWB) bandwidths. The techniques include identifying a context or environment in which audio data is captured, and adaptively changing a level of noise suppression applied to the audio data prior to bandwidth compressing (e.g., encoding) based on the context. For a valid speech context, an audio pre-processor may set a first level of noise suppression that is relatively aggressive in order to suppress noise (including music) in the speech signals. For a valid music context, the audio pre-processor may set a second level of noise suppression that is less aggressive in order to leave the music signals undistorted. In this way, a vocoder at a transmitter side wireless communication device may properly encode both speech and music signals with minimal distortions.