Frequency-Lowered Speech Processing for Sibilant Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hearing aid devices using positive rank scaling techniques often increase confusion between sibilant fricatives (/s/ and /∫/), as frequency-lowered /s/ is often perceived as /∫/, due to a nonmonotonic relationship between frequency-lowered spectral peaks and perception.
Innovation Solution
A sound processing device system that performs negative rank ordering of frequencies, inverting the ordering of high-frequency portions of the input audio signal to low-frequency portions in the output, thereby preserving the discrimination between /s/ and /∫/ by applying a reciprocal function to frequency bands, ensuring that high frequencies become low frequencies and mid-range frequencies remain unchanged.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If positive rank scaling frequency lowering is applied, then high-frequency speech information becomes audible, but discrimination between sibilant fricatives /s/ and /∫/ deteriorates
Solution Approach 1:
The patent applies negative rank scaling, which inverts the frequency mapping relationship. Instead of mapping high frequencies to low frequencies in ascending order (positive rank scaling), it maps them in descending order. This inversion preserves the spectral contrast between /s/ and /∫/ by reversing the rank ordering, thereby maintaining discrimination while achieving audibility.
Solution Approach 2:
The patent changes the scaling parameter from positive to negative rank scaling. This parameter change fundamentally alters the frequency mapping function, transforming how spectral information is compressed or transposed while preserving the relative discrimination between different fricative sounds.
2Reliability
If frequency compression is applied to fit the entire frequency range into the audible region, then more speech information becomes audible, but frequency resolution and sound discrimination worsen
Solution Approach 1:
By using negative rank scaling instead of positive rank scaling, the patent inverts the compression mapping. This inversion allows the frequency range to be compressed into the audible region while preserving frequency resolution through the reversed rank ordering, thereby maintaining sound discrimination.
3Reliability
If spectral features are transposed to lower frequencies, then speech information in the inaudible region becomes audible, but confusion between similar sounds increases
Solution Approach 1:
The patent applies negative rank scaling to transpose spectral features, which inverts the mapping relationship. This inversion preserves the distinctive spectral characteristics of different sounds by reversing the rank order, thereby reducing confusion between similar sounds while making inaudible speech information audible.
Data Source
AI summary
Among other things, a sound processing device system is disclosed to assist a hearing-impaired human listener recognize speech sounds or phonemes. The device system may be configured at least to generate an output audio signal at least by transposing and causing a negative rank ordering of frequency of at least a portion of the input audio signal. Compression also may be performed on the at least the portion of the input audio signal as part of generating the output audio signal. The negative rank ordering may be performed on a high-frequency portion of the input audio signal that becomes a low-frequency portion of the output audio signal by the transposing. The low-frequency portion of the output audio signal may represent an inverted ordering of frequencies or frequency segments present in the high-frequency portion of the input audio signal.