Hearing Aid Speech Correction via Phoneme Label Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing synthesized speech systems for hearing aids cause discomfort due to excessive manipulation of voice waveforms and lack of standards in consonant and vowel characteristics, leading to confusion in phoneme perception, even for individuals with normal hearing.
Innovation Solution
A method for correcting a synthesized speech set for hearing aids by correlating phoneme labels with synthesized speech data based on user feedback from a testing process, allowing for the adjustment of phoneme labels to improve recognition and reduce discomfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If emphasis processing is applied to synthesized speech to improve recognition accuracy, then phoneme recognition is improved, but user discomfort increases due to excessive manipulation of amplitude and frequency
Solution Approach 1:
The patent implements dynamic adjustment of emphasis processing parameters based on individual user auditory characteristics. The system adapts the degree of amplitude and frequency manipulation according to each user's specific hearing profile, transforming the static emphasis processing into a dynamic, user-specific process that maintains recognition accuracy while reducing discomfort.
Solution Approach 2:
The patent changes the parameters of emphasis processing (amplitude, frequency, duration) based on measured auditory characteristics of individual users. By adjusting these parameters according to user-specific hearing thresholds and sensitivities, the system optimizes the balance between phoneme recognition and user comfort.
2Adaptability or versatility
If standardized characteristics for consonants and vowels are not established in synthesized speech, then system compatibility is maintained, but phoneme perception confusion occurs even for normal hearing individuals
Solution Approach 1:
The patent establishes specific acoustic characteristics for different phoneme types (consonants, vowels, semi-vowels) with localized quality parameters. Each phoneme category has defined characteristics such as duration, amplitude, and frequency ranges that distinguish it from others, enabling clear phoneme perception while maintaining system compatibility through standardized local properties.
Solution Approach 2:
The patent performs preliminary analysis and classification of phonemes into categories (consonant, vowel, semi-vowel) before synthesis, applying predetermined characteristic parameters to each category. This preliminary organization ensures consistent phoneme perception across different synthesized speech outputs while maintaining adaptability.
3Reliability
If synthesized speech data is modified to improve user comfort and recognition, then speech quality is improved, but the complexity of data processing and correction increases
Solution Approach 1:
The patent segments synthesized speech data into discrete phoneme units and applies correction rules at the phoneme level rather than processing entire speech sequences. This segmentation allows for simpler, more manageable correction operations on individual phonemes while maintaining overall speech quality and recognition reliability.
Solution Approach 2:
The patent creates corrected versions of synthesized speech data by copying and modifying phoneme characteristics according to established rules, rather than completely regenerating the speech data. This copying approach simplifies processing by working with replicated phoneme units that can be systematically adjusted.
Data Source
AI summary
A method for correcting a synthesized speech set for hearing aid according to an aspect of the present invention includes the steps of outputting first synthesized speech for testing on the basis of first synthesized speech data for testing correlated with a first phoneme label in a synthesized speech set for testing, accepting a first answer selected by a user, outputting second synthesized speech for testing on the basis of second synthesized speech data for testing correlated with a second phoneme label in the synthesized speech set for testing, accepting a second answer selected by the user, and correlating first synthesized speech data for hearing aid with the second phoneme label instead of second synthesized speech data for hearing aid in a synthesized speech set for hearing aid, in a case in which the first answer matches the second phoneme label and also the second answer does not match the second phoneme label.