Formant-Aware Spectral Analysis of Sustained Vowels of Pathological Breathy Voice

preprint OA: closed CC-BY-4.0
📄 Open PDF View at publisher

Abstract

Objectives: This paper reports the effectiveness of formant-aware spectral parameters to predict the perceptual breathiness rating. Breathy voice has a higher first harmonic, steeper spectral slope, and higher turbulent noise than normal voice. Measuring spectral parameters of acoustic signal over formant regions is a known approach to capture the properties related to breathiness. This study examines this approach by testing the contemporary spectral parameters and algorithms within the framework, alternate frequency band designs, and vowel effects. Methods: . Sustained vowel recordings (/a/, /i/, and /u/) of speakers with voice disorders in the German Saarbrueken Voice Database were considered (n=368). Recordings with spectral irregularity or roughness perception were excluded from the study. Four speech language pathologists perceptually rated the recordings for breathiness on a 100-point scale, and their averages were used as the ratings of the recordings. The acoustic spectra were segmented into four frequency bands according to the vowel formant structures. Five different spectral parameters were considered in each band or between bands (13 total plus the fundamental frequency) to predict the perceptual breathiness rating. Results: . Linear combinations of spectral parameters, led by the formant-focused harmonics-to-noise ratios (HNRs), were shown to explain up to 85% of the variance in perceptual breathiness ratings of disordered voice. This performance exceeded that of the Acoustic Breathiness Index (82%). Also, the best performing parameter (the HNR over the first two formants, 78%) explained more variances in the breathiness than the smoothed cepstrum peak prominence (74%). Some vowel effects were observed in the perceptual rating (higher for /u/), in predictability (5% lower for /u/), and in model parameter selections. Conclusions: . Strong breathiness correlates were found by segmenting the spectrum to isolate the portion most affected by breathiness.

My notes (saved in your browser only)

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. The paper's references may be in our DB but unresolved to ``paper_id`` (resolution happens at ingest when the cited DOI matches a row we already have). Run the cross-source citation reconcile pass to retry.

Source provenance

europepmc
last seen: 2026-05-19T01:45:01.086888+00:00
unpaywall
last seen: 2026-06-06T02:00:05.402940+00:00
License: CC-BY-4.0