Hearing Aid Frequency Compression Knee Point Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hearing aid frequency compression algorithms lack a defined strategy for parameterization to enhance speech intelligibility, particularly failing to account for the fine acoustic structure of consonants and vowels, leading to laborious and time-consuming manual fine-tuning for improved results.
Innovation Solution
A method to determine the knee point of the frequency compression characteristic curve based on the maximum audible frequency of the user, using the Bark scale, which allows for optimized information transmission and minimizes frequency compression below 1.5 kHz to distinguish between male and female voices, with compression rates up to 4:1 above this point.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If frequency compression algorithm is applied to make high-frequency information audible, then speech intelligibility can be improved, but the algorithm lacks defined parameterization strategy leading to laborious manual fine-tuning
Solution Approach 1:
The patent applies parameter changes by systematically varying frequency compression parameters (compression ratio, knee point frequency, transition bandwidth) based on the user's audiogram and speech spectrum characteristics. This allows automatic optimization of speech intelligibility without requiring complex manual fine-tuning, directly resolving the contradiction between improving speech understanding and reducing parameterization complexity
Solution Approach 2:
The patent implements self-service through an automatic fitting procedure that computes frequency compression parameters based on objective measurements (audiogram, speech spectrum) without requiring expert manual adjustment. The system self-optimizes by analyzing the user's specific hearing loss pattern and automatically determining optimal compression settings, eliminating the need for laborious manual fine-tuning while maintaining reliable speech intelligibility improvement
2Reliability
If manual fine-tuning is performed to improve speech intelligibility, then speech recognition can be enhanced, but the process becomes extremely laborious and time-consuming
Solution Approach 1:
The patent applies preliminary action by pre-computing optimal frequency compression parameters based on the user's audiogram and speech spectrum characteristics before actual use. The automatic fitting procedure performs all necessary optimizations in advance, eliminating the need for time-consuming manual fine-tuning during fitting sessions. This preliminary computation of parameters directly reduces adjustment time while maintaining enhanced speech recognition
Solution Approach 2:
The patent substitutes the mechanical manual adjustment process with an automated computational system. Instead of requiring audiologists to manually tweak parameters, the system uses algorithms to automatically compute optimal settings based on objective measurements. This replacement of manual mechanical adjustment with automated computation dramatically reduces the time required while achieving the same or better speech recognition outcomes
3Reliability
If frequency compression maps high frequencies to lower frequencies, then high-frequency information becomes audible, but fine acoustic structure of consonants and vowels may be lost
Solution Approach 1:
The patent applies local quality by implementing frequency-dependent compression ratios and selective frequency mapping. Different frequency regions are compressed with different ratios, and specific formant regions are preserved with minimal compression to maintain the fine acoustic structure of consonants and vowels. This localized differentiation allows high-frequency information to become audible while preserving critical speech discrimination cues
Solution Approach 2:
The patent uses formant preservation as an intermediary mechanism to bridge between high-frequency compression and fine acoustic structure preservation. By identifying and protecting formant regions (critical for vowel and consonant discrimination) while compressing other high-frequency regions, the system acts as an intermediary that allows frequency compression to improve audibility without losing essential speech information
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The method involves determining (15) a maximum audible frequency of a hearing device. The knee point is determined by predefined rule in dependence on the maximum audible frequency. The parameters of frequency compression characteristic are constituted on a basis of frequency groups such that number of frequency groups is determined based on the maximum audible frequency. The maximum compression rate is set to be 4. An independent claim is included for method for adjusting a binaural hearing system.