Assessment of the possibility of acoustic voice analysis

Introduction . Voice disorders occur in approximately 30% of the country’s population. The most studied characteristics of the voice include fundamental frequency, pitch and amplitude, harmonic-to-noise ratio, cepstral peak severity, acoustic quality index of voice, maximum phonation time, variations in fundamental frequency and number of pauses in speech signals. Aim . Literature review assessing the possibility of acoustic voice analysis in patients with dysphonia. Materials and methods . The authors searched for publications in the electronic databases PubMed, Web of Science, Google Scholar and ELibrary. The search was carried out using the following keywords: “voice acoustic analysis”, “voice disorder”, “artificial neural network”, “dysphonia”, “standard deviation of fundamental frequency”, “voice quality”, “acoustic voice analysis”. Results and discussion . Fundamental frequency may be more sensitive to objective clinical assessment of voice than pitch and amplitude. The severity of the cepstral peak is an integral part of the acoustic analysis of the voice, helping to determine the differences between dysphonic and normal voices. Cepstral analysis is more sensitive to subtle dysphonic changes than vowel analysis methods. Despite the high analytical accuracy, ease of use of machine learning, as well as the promise of this approach in the diagnosis of dysphonia, the clinical application of this technology requires further research Conclusions . Acoustic Analysis of Voice offers numerous advantages such as non-invasiveness, cost-effectiveness, and ease of use, facilitating the acquisition of objective data for evaluating the severity of voice disorders and serving as an indispensable tool for identifying pathologies associated with phonation disturbances. According to the literature, the most informative Acoustic Analysis of Voice parameters include fundamental frequency metrics, pitch and amplitude indices, cepstral peak prominence, voice quality index, maximum phonation time, and the relative noise level in the speech signal.