Subband Audio Compression for Clearer Hearing Aid Replay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional digital signal processing (DSP) methods for sound personalization in hearing aids fail to provide an enhanced listening experience for individuals with hearing impairments due to poor frequency resolution and temporal smearing, leading to discomfort and difficulty in noisy environments.
Innovation Solution
The development of biologically-inspired DSP algorithms that mimic the functional processing of the healthy human ear, involving spectral decomposition, feedforward and feedback dynamic range compression, and phase-linear filtering to control distortion and improve sound fidelity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If equalization techniques are used to amplify frequencies according to a user's hearing profile, then the user can recapture previously unheard frequencies, but the user experiences loudness discomfort and hearing damage risk
Solution Approach 1:
The patent applies dynamic range compression that adapts to the input signal characteristics and listening context. The system dynamically adjusts compression ratios and thresholds based on the audio content being played, allowing quiet sounds to be amplified while preventing loud sounds from causing discomfort. This dynamic adaptation resolves the contradiction by making the amplification selective rather than static across all frequencies and volumes.
Solution Approach 2:
The system changes multiple parameters simultaneously including compression threshold, compression ratio, and attack/release times based on the audiogram data and signal characteristics. By adjusting these parameters dynamically rather than applying fixed equalization gain, the system can amplify frequencies needed by the user while preventing harmful loudness, thus resolving the contradiction between improved frequency detection and reduced loudness discomfort.
2Measurement precision
If dynamic range compression is used to amplify quieter sounds, then the dynamic range is narrowed, but low frequency rumble prevents amplification of high frequency sounds of interest
Solution Approach 1:
The patent segments the audio signal into multiple frequency bands and applies independent dynamic range compression to each band. This allows the system to compress quiet sounds in high frequency bands while leaving low frequency rumble relatively unaffected, preventing the rumble from masking the amplified high frequency sounds. The segmentation enables selective processing that resolves the contradiction between amplifying quiet sounds and preserving high frequency information.
Solution Approach 2:
The system applies different compression characteristics to different frequency regions based on the user's specific hearing loss profile. Frequency bands where the user has greatest loss receive more aggressive compression, while bands with less loss receive minimal processing. This local customization ensures that quiet high frequency sounds are amplified without being masked by low frequency rumble, resolving the contradiction through spatially differentiated processing.
3Adaptability or versatility
If conventional hearing aid processing is used for real world situations, then the listener can detect faint sounds and understand loud speech, but the processing is not optimized for audio content on mobile devices
Solution Approach 1:
The system dynamically adapts its processing based on the type of audio content being played. When detecting music or recorded audio versus live speech or environmental sounds, the system adjusts compression parameters, frequency weighting, and processing intensity accordingly. This dynamic adaptation allows optimization for mobile device audio content while maintaining effectiveness for real-world situations, resolving the contradiction between versatility and precision.
Solution Approach 2:
The system changes processing parameters based on the audio source and listening context. For mobile device audio content, the system applies different compression ratios, frequency responses, and signal processing algorithms compared to live hearing aid situations. This parameter adaptation enables the same device to provide optimized listening experience for both real-world sounds and recorded audio content, resolving the contradiction between adaptability and measurement precision.
Data Source
AI summary
Systems and methods for processing an audio signal are provided for replay on an audio device. An audio signal is spectrally decomposed into a plurality of subband signals using band pass filters. Each of the subband signals are provided to a respective modulator and subsequently, from the modulator output, provided to a respective first processing path that includes a first dynamic range compressor, DRC. Each subband signal is feedforward compressed by the respective first DRC to obtain a feedforward-compressed subband signal, wherein the first DRC is slowed relative to an instantaneous DRC. Subsequently, each feedforward-compressed subband signal is provided to a second processing path that includes a second DRC, wherein the feedforward-compressed subband signal is compressed by the respective second DRC and outputted to the respective modulator. Modulation of the subband signals is then performed in dependence on the output of the second processing path. Finally, the feedforward-compressed subband signals are recombined.