Multiband Audio Compression Using Cochlear Suppression Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio compression systems for hearing aids and implantable auditory prostheses lack effective compression functionality to restore normal cochlear two-tone suppression and loudness growth, leading to reduced spectral contrasts and inadequate loudness perception in hearing-impaired individuals.
Innovation Solution
A multiband audio compression system that separates audio signals into frequency bands, applies time-varying gains based on suppressive levels calculated using equations modeling normal two-tone suppression, and combines the bands to produce an output, simulating the cochlear processes of frequency separation, dynamic-range compression, and amplification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional audio compression systems are used, then device complexity is reduced, but spectral contrasts are degraded and loudness perception is inadequate
Solution Approach 1:
The audio signal is divided into multiple frequency bands using a filter bank, allowing independent processing of each band. This segmentation enables the system to apply frequency-specific compression parameters that preserve spectral contrasts while managing complexity through modular band processing
Solution Approach 2:
Different compression parameters (thresholds, ratios, attack/release times) are applied to different frequency bands based on their specific characteristics and the user's hearing profile. This local quality approach optimizes loudness perception in each band without requiring uniformly complex processing across all frequencies
2Reliability
If compression functionality is added to restore cochlear suppression, then loudness perception is improved, but device complexity increases
Solution Approach 1:
Compression parameters including thresholds and ratios are pre-calculated based on the user's audiometric data and fitted hearing profile before actual audio processing. This preliminary configuration reduces real-time computational complexity while maintaining reliable loudness perception tailored to individual needs
Solution Approach 2:
The system continuously monitors the compressed output across frequency bands and dynamically adjusts compression parameters to maintain natural loudness growth and simulate cochlear suppression. This feedback mechanism improves loudness perception while managing complexity through adaptive rather than purely algorithmic processing
Data Source
AI summary
Method and systems (e.g., for use in hearing aids, IAPs, etc) to process an audio signal include receiving an audio signal input; separating the audio signal input into a plurality of frequency bands; and compressing each of the plurality of frequency bands. Compressing each respective frequency band of the plurality of frequency bands may include applying a time-varying gain to each respective frequency band based on a suppressive level for the respective frequency band resulting in a compressed respective frequency band (e.g., wherein the suppressive level for the respective frequency band is dependent on the audio signal input level of one or more frequency bands adjacent to the respective frequency band to which the gain is applied). The compressed respective frequency bands may be combined for use in providing an audio signal.