Voice Call Noise Suppression With User-Selectable Background Audio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise suppression algorithms in voice communications do not allow users to selectively suppress or include specific types of background noise, leading to frustration in noisy environments and missed important sounds.
Innovation Solution
A voice communication device that classifies background noise types and allows users to select which types to include or exclude, using an audio context detector and noise suppressor to customize noise suppression based on user preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If noise suppression algorithms automatically suppress all background audio data, then noise reduction is improved, but user control and transmission of important sounds deteriorate
Solution Approach 1:
The patent segments background audio data into multiple types (e.g., noise, music, speech, environmental sounds) and applies different suppression decisions to each type. The noise suppressor can selectively suppress certain types while preserving others, allowing users to control which background sounds are transmitted based on their importance or preference.
2Measurement precision
If noise suppression algorithms suppress all background noise, then speech clarity is improved, but important background sounds are lost
Solution Approach 1:
The patent applies different quality levels of noise suppression to different types of background audio data. Critical sounds (e.g., alarms, speech) are preserved with minimal suppression, while non-critical noise is heavily suppressed. This local differentiation maintains speech clarity while preserving important background information.
3Reliability
If user-selectable noise suppression is implemented, then communication quality is improved, but device complexity increases
Solution Approach 1:
The patent implements a dynamic noise suppression system where the suppression level and type are adjusted in real-time based on user selections and detected audio conditions. The system can adaptively change which background audio types are suppressed during the communication, providing high communication quality without requiring a completely static complex system.
Data Source
AI summary
An apparatus for audio communication includes a memory configured to receive audio data from a user of a voice communication, and one or more processors in communication with the memory. The one or more processors are configured to receive the audio data for the voice communication, which includes voice data of the user and background audio data. The processors are further configured to classify the background audio data into a plurality of types of background audio data, determine to not suppress a subset of the plurality of types of background audio data, process the audio data to not suppress the subset of the plurality of types of background audio data to generate output audio data, and transmit the output audio data.