A Hearing Test Simulator GUI for clinical testing of speech recognition

Area of Science:

• Audiology
• Signal Processing
• Human-Computer Interaction

Background:

• Assessing speech recognition in noise is crucial for understanding auditory perception.
• Existing methods for testing in simulated noisy environments can be costly and complex.
• A need exists for accessible tools for evaluating speech recognition performance under varying signal-to-noise ratios (SNR).

Purpose of the Study:

• To present a MATLAB-based Graphical User Interface (GUI) for speech recognition testing.
• To evaluate subjects' speech perception abilities in diverse noisy environments.
• To provide a cost-effective and reliable simulation for clinical and engineering evaluations.

Main Methods:

• Development of a MATLAB-based GUI for presenting auditory stimuli.
• Recording and analyzing subjects' word perception responses.
• Collecting test data under controlled conditions with varying noise types (babble, traffic, machinery, white noise) and SNRs.
• Saving word recognition rates and scores into a database for analysis.

Main Results:

• The GUI enables accurate identification of correctly perceived words.
• Word recognition rates can be systematically recorded across different noise conditions and SNRs.
• Test data and scores are stored, facilitating repeated testing cycles and subsequent analysis.
• The system provides a reliable and cost-effective method for simulating real-world auditory challenges.

Conclusions:

• The developed MATLAB GUI offers a valuable tool for speech recognition testing in noisy environments.
• It supports both clinical evaluation of auditory function and engineering research in signal processing.
• The software provides a reliable, cost-effective, and flexible platform for assessing speech perception under controlled adverse conditions.