Smart Diagnostic Tool for Voice Pathologies Using Audio Signal Processing

Abstract

Voice disorders are a lot more common than people think, and catching them early makes a big difference in treatment outcomes. The problem is that most standard diagnostic methods — like laryngoscopy — are invasive, require hospital equipment, and are not practical for regular screening. This paper presents a simple, non-invasive diagnostic tool built on LabVIEW that uses basic audio signal processing to detect voice pathologies in real time. The system records voice samples from both healthy individuals and patients with diagnosed vocal conditions, then analyses them using two straightforward techniques: amplitude-time waveform analysis and power spectral density (PSD). Healthy voices showed smooth, regular waveforms with clear harmonic patterns in the frequency domain, while pathological voices showed irregular waveforms and distorted spectra. Three acoustic markers were also extracted — jitter (0.0237), shimmer (0.1442), and harmonic-to-noise ratio (13.56 dB) — all of which fell outside the normal clinical range for pathological samples. The results confirm that this tool can clearly tell apart healthy and disordered voices without any invasive procedure. It is designed to be affordable, easy to use, and practically deployable in clinics, research labs, or even remote healthcare setups.

Introduction

Voice disorders such as dysphonia, vocal nodules, and laryngeal cancer significantly affect speech and may lead to serious health issues if not diagnosed early. Traditional diagnostic methods like laryngoscopy are invasive, expensive, and unsuitable for large-scale or remote screening. To address this, the study proposes a non-invasive alternative using audio signal processing and a LabVIEW-based system that analyzes recorded voice signals instead of using clinical imaging.

The proposed tool processes voice recordings in both time and frequency domains to identify pathological patterns. It extracts acoustic features such as jitter, shimmer, harmonic-to-noise ratio (HNR), amplitude-time waveforms, and power spectral density (PSD). These features help distinguish healthy voices from disordered ones by revealing irregular vibrations and spectral noise associated with vocal fold abnormalities. The system provides a side-by-side visual comparison of healthy and pathological voices for easier clinical interpretation.

A dataset of healthy individuals and patients with vocal disorders (including dysphonia and in-situ carcinoma) was collected using controlled recordings of sustained vowels and continuous speech. The system pipeline includes voice acquisition, preprocessing (noise reduction, normalization, segmentation), feature extraction, and visualization, all implemented in LabVIEW for real-time operation.

Preprocessing improves signal quality by filtering noise, standardizing amplitude, and isolating relevant speech segments. Feature extraction then analyzes waveform behavior and frequency patterns using FFT-based spectral analysis. Healthy voices show stable periodic waveforms and clear harmonic structures, while pathological voices exhibit irregular patterns and noisy, distorted spectra.

Conclusion

This paper presented a LabVIEW-based tool for detecting voice pathologies using straightforward audio signal processing — no invasive procedures, no expensive equipment, no complex machine learning. The system analyses voice recordings through amplitude-time waveforms and PSD plots, and backs those up with three acoustic markers: jitter, shimmer, and HNR. Testing on samples from healthy individuals and patients with in-situ carcinoma showed clear, consistent differences between the two groups. All three acoustic markers were outside normal ranges for pathological samples, and this matched what the waveforms and spectral plots were showing visually. The system processes each recording in a matter of seconds, works with both live microphone input and pre-recorded files, and produces output that is easy to read without any technical background in signal processing. The goal of this tool is to give ENT specialists, speech therapists, and general clinicians a practical, affordable first line of assessment — especially in settings where standard diagnostic equipment is not available. With further development, including validation on larger datasets and integration with portable hardware, this system has the potential to become a genuinely useful tool in everyday voice healthcare.

References

[1] D. Gurdak, K. Pancerz, J. Szkola, and J. Warchol, \"Computer-Aided Diagnosis of Laryngopathies in the LabVIEW Environment: Exemplary Implementation,\" in Advances in Intelligent Analysis of Medical Data and Decision Support Systems, R. Kountchev and B. Iantovics, Eds., Studies in Computational Intelligence, vol. 473. Cham, Switzerland: Springer, 2013, pp. 157-167. [2] A. S. Amrutha, A. S. Nair, C. J. Mathew, J. E. George, and M. George, \"Characterization of Speech Signal in LabVIEW Platform,\" Int. J. Eng. Res. Technol., vol. 3, no. 5, pp. 1-4, 2015. [3] R. Martinek, \"Speech Signal Processing Using Microphones NI-9234 and LabVIEW,\" in Proc. Int. Conf. Computer Modeling and Simulation, 2018, pp. 112-117. [4] National Instruments, Signal Analysis -- LabVIEW API Reference, Austin, TX, USA: National Instruments, 2025. [5] National Instruments, LabVIEW Jitter Analysis Toolkit -- Overview, Austin, TX, USA: National Instruments, 2014. [6] A. Banacia, \"A Learning Module on Sound Analysis Using LabVIEW,\" in Proc. IEEE Int. Conf. Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM), 2015, pp. 1-5. [7] R.-C. V. Todorov, \"Audio Signal Processing System Design Using LabVIEW,\" in Proc. Int. Conf. Electronics, Signal Processing and Communication Systems (ICEST), 2015, pp. 225-229. [8] S. K. Singla and R. K. Yadav, \"Optical Character Recognition Based Speech Synthesis System Using LabVIEW,\" J. Appl. Res. Technol., vol. 12, no. 5, pp. 843-850, 2014. [9] National Instruments, LabVIEW Advanced Signal Processing Toolkit User Manual, Austin, TX, USA: National Instruments, 2025. [10] National Instruments, \"Using LabVIEW for Voice Signal Analysis,\" NI Community Technical Report, Austin, TX, USA: National Instruments, 2008.

Copyright

Copyright © 2026 Riya Singh, Dr. Lochan Jolly . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.