Receiving-Side Speech Quality Detection for Selective Noise Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise suppression techniques in device communication degrade speech quality and waste processing resources by applying unnecessary noise suppression to already degraded speech data.
Innovation Solution
Implement a speech quality predictor and noise detector on the receiving device to determine if noise suppression is needed, selectively applying it only when speech quality is above a threshold and noise is above a certain level, thereby omitting it when it would degrade speech quality further.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If noise suppression is applied to speech data, then noise is reduced, but speech quality degrades due to processing by components in the transmission chain
Solution Approach 1:
The patent implements dynamic selection of noise suppression processing based on real-time assessment of speech data quality and noise characteristics. The system adapts its processing approach by conditionally applying or omitting noise suppression algorithms, transitioning between different processing states rather than using a fixed processing pipeline. This dynamic adaptation resolves the contradiction by applying noise suppression only when speech quality permits and noise levels warrant intervention.
Solution Approach 2:
The system changes the processing parameter (noise suppression application) based on assessed conditions of the speech data. By monitoring speech quality metrics and noise characteristics, the system adjusts whether noise suppression is applied, effectively changing the processing state from always-on to conditionally-applied. This parameter change allows optimization of both noise reduction and speech quality preservation.
2Object-affected harmful factors
If noise suppression is applied to all speech data, then noise is suppressed, but processing resources and power consumption increase unnecessarily
Solution Approach 1:
The patent applies partial action by selectively applying noise suppression only to portions of speech data that meet specific criteria (above threshold quality and sufficient noise level). Rather than applying noise suppression universally to all speech data, the system performs partial processing based on conditional assessment, thereby reducing unnecessary power consumption while maintaining effective noise suppression where needed.
Solution Approach 2:
The system performs self-assessment of speech data quality and noise characteristics using integrated analysis components, then makes autonomous decisions about whether noise suppression is required. This self-service capability eliminates the need for continuous noise suppression processing, allowing the system to conserve power by independently determining when processing is actually beneficial.
3Object-affected harmful factors
If noise suppression is applied to degraded speech data, then noise is reduced, but speech quality degrades further due to additional processing
Solution Approach 1:
The patent implements a feedback mechanism where speech data is first assessed for quality and noise characteristics before noise suppression is conditionally applied. This feedback loop allows the system to identify degraded speech data and exclude it from further noise suppression processing, preventing additional quality degradation. The feedback-based decision-making ensures that noise suppression is applied only when it will not harm speech quality.
Data Source
AI summary
Implementations described herein relate to selective noise suppression for speech data in device communication. In some implementations, a computer-implemented method includes receiving, by a receiving device, speech data over a communication link from a transmitting device, where the speech data is processed by one or more components of the transmitting device or the communication link. A quality of the speech data is determined using a speech quality predictor implemented on the receiving device. In response to determining that the quality of the speech data is above a threshold quality and/or includes significant noise, noise suppression is provided for the speech data using a noise suppressor implemented on the receiving device. In response to determining that the quality of the speech data is below the threshold quality, the noise suppression is omitted for the speech data by the receiving device.