Selective Hearing System Using Audio Source Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hearing devices, including binaural systems, struggle to effectively isolate and enhance specific sound sources in complex auditory environments, leading to poor speech comprehensibility and increased noise interference, especially in scenarios with multiple speakers or background noise.
Innovation Solution
A system that utilizes multiple microphones to detect and classify audio sources, assigning position information and modifying audio signals based on type, generating binaural room impulse responses to enhance desired sounds and reduce undesired noise, allowing for intelligent control of sound isolation and emphasis on specific sound sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional hearing devices process signals independently for each ear, then device complexity is reduced, but speech comprehensibility deteriorates due to loss of spatial information
Solution Approach 1:
The patent segments the audio signal processing into distinct functional modules: detection module for identifying sound sources, classification module for categorizing sound types, and modification module for selective signal enhancement. This segmentation allows independent processing of spatial and spectral information while maintaining overall system complexity at acceptable levels.
Solution Approach 2:
The patent transitions from monaural to binaural processing by adding the spatial dimension. It utilizes interaural time differences, interaural level differences, and spectral cues across both ears to create a three-dimensional auditory representation, thereby preserving spatial information that would be lost in single-ear processing.
2Loss of information
If binaural hearing devices couple correction factors between ears, then spatial information is preserved, but speech comprehensibility deteriorates in complex environments due to uniform amplification of all sounds
Solution Approach 1:
The patent applies different processing qualities to different spatial locations and sound sources. Desired sound sources receive enhancement processing while undesired sounds receive suppression or leave unchanged. This localized quality adjustment allows selective improvement of speech comprehensibility without uniformly amplifying all sounds, thereby reducing noise interference while preserving spatial information.
Solution Approach 2:
The patent dynamically changes multiple parameters including gain factors, spatial position, and spectral characteristics based on the detected sound source properties and user preferences. By adjusting these parameters selectively for different sound sources, the system enhances desired speech while suppressing background noise, resolving the contradiction between preserving spatial information and reducing noise interference.
3Loss of information
If multiple microphones are used for sound source detection, then speech comprehensibility improves through better spatial filtering, but device complexity increases due to beamforming computations
Solution Approach 1:
The patent performs preliminary detection and classification of sound sources before applying complex beamforming operations. By first identifying potential speech sources and their spatial locations using simpler algorithms, the system prepares targeted processing regions, thereby reducing the overall computational burden of beamforming while maintaining speech signal quality.
Solution Approach 2:
The patent applies beamforming computations selectively rather than uniformly across all frequency ranges and time periods. It focuses computational resources on critical speech frequencies and time segments where speech signals are present, performing partial beamforming operations only where necessary. This approach maintains speech comprehensibility while significantly reducing overall device complexity.
4Object-affected harmful factors
If hands-free devices suppress non-speech signals, then noise interference is reduced, but important environmental information is lost
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously monitors the processed audio output and adjusts suppression levels accordingly. When important environmental sounds occur (such as alarms or significant ambient events), the feedback loop detects these and reduces suppression, allowing the information to pass through. This resolves the contradiction by dynamically balancing noise suppression with environmental information preservation based on real-time conditions.
Solution Approach 2:
The patent transforms static suppression thresholds into dynamic, adaptive parameters that change based on the acoustic environment, user activity, and detected sound source characteristics. The suppression level is not fixed but adjusts in real-time, being more aggressive during quiet periods and more permissive when environmental information becomes important, thereby balancing noise reduction with information preservation.
Data Source
AI summary
A system and method for assisting selective hearing includes a detector for detecting an audio source signal portion of one or more audio sources by using at least two received microphone signals of a hearing environment. A position determiner allocates position information to each of the one or more audio sources. An audio type classifier assigns an audio source signal type to the audio source signal portion of each of the one or more audio sources. A signal portion modifier varies the audio source signal portion of at least one audio source of the one or more audio sources depending on the audio signal type of the audio source signal portion of the at least one audio source so as to obtain a modified audio signal portion of the at least one audio source. The system includes a signal generator.