Adaptive Frequency Transposition for Hearing Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing frequency transposition methods for hearing devices distort vowels and non-fricative sounds when the cut-off frequency is lowered below 1,500 Hz, leading to poor sound quality and increased confusion for users with high-frequency hearing loss, and are not effective for transposing music due to pitch distortions.
Innovation Solution
A frequency transposition scheme that adaptively selects signal components based on momentary characteristics, such as auditory expectation and energy distribution, using a frequency stacking algorithm to transpose signal components from source stacks to destination stacks while preserving spectral contrast and minimizing distortion, and applying pre- and post-weighting functions to optimize sound quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the cut-off frequency is lowered below 1,500 Hz to extend the frequency transposition range, then more high-frequency hearing loss can be compensated, but vowel distortion and sound quality degradation occur
Solution Approach 1:
The patent applies different processing strategies to different frequency regions and signal types. Vowels below the cut-off frequency are preserved without transposition to maintain quality, while consonants and fricatives above the cut-off frequency are transposed to compensate for hearing loss. This localized quality approach allows the system to maintain high sound quality in critical regions while extending transposition capability in other regions.
Solution Approach 2:
The system dynamically adjusts the cut-off frequency threshold based on the instantaneous spectral characteristics of the input signal. By analyzing the energy distribution and identifying significant spectral segments in real-time, the system adapts the transposition parameters to preserve vowels when present while transposing consonants and fricatives, thereby maintaining sound quality while maximizing the beneficial transposition range.
2Reliability
If frequency transposition is applied to expand bandwidth for hearing impaired users, then speech intelligibility can be improved, but confusion and altered pitch perception increase
Solution Approach 1:
The patent carefully controls the transposition parameters including the cut-off frequency threshold, transposition ratio, and frequency mapping function. By optimizing these parameters, the system transposes high-frequency consonants and fricatives to lower frequencies where they can be heard, improving speech intelligibility while maintaining natural pitch relationships and minimizing confusion through controlled parameter selection.
Solution Approach 2:
The system acts as an intermediary that selectively transposes only the necessary frequency components (consonants and fricatives) while leaving vowels and low-frequency components unchanged. This selective transposition approach maintains the natural pitch perception of the original signal while providing the bandwidth expansion needed for hearing impaired users, thereby reducing confusion.
3Use of energy by moving object
If high gain amplification is applied to compensate for high frequency hearing loss, then audibility can be improved, but feedback problems and limited bandwidth constrain the solution
Solution Approach 1:
The patent segments the frequency spectrum into multiple bands and applies frequency transposition to map high-frequency components to lower frequency bands. This segmentation approach allows the system to provide effective amplification for high-frequency hearing loss without the feedback problems associated with high gain amplification in the original high-frequency range, as the transposed signals occupy different frequency spaces.
Data Source
Figure 1~3
Figure 4~6
Figure 7~9
AI summary
A method for operating a hearing device by applying a frequency transposition scheme to an input signal of the hearing device comprising an input transducer, a signal processing unit and an output transducer, the method comprising the steps of transforming the input signal from time domain into frequency domain by applying a transformation function in order to obtain an input spectrum having a frequency range comprising a source region (20) and a destination region (30), adaptively selecting signal components of the source region (20) taking into account momentary characteristics of the input signal, transposing the selected signal components to the destination region (30), and supplying the output spectrum or a transformation thereof to the output transducer, the output spectrum comprising signal components of the destination region (30).