[Variability in the digital voice analysis depending on the analyzed vocal, in normal patients and in patients with...

The aim of this study is to decide which of Spanish's 5 vowels is the most suitable for the analysis of the voice by means of the program Dr. Speech 3.0. We assessed 98 patients: 51 normal subjects and 47 hoarse patients (34 polyps, 7 nodules, 2 hyperfunctional dysphonia, 4 other pathologies) before and after following a surgical and/or speech therapy. The methodology of the study included: protocolized questionnaire, ENT examination, perceptual evaluation of the hoarse voices and physical analysis of the 5 sustained vowels. Jitter, shimmer, NNE and HNR in the voice varied depending on the analyzed vowel and from normal subjects to hoarse patients. NNE and HNR for vowels [o] and [u] was higher than for [a], such differences were statistically significant (p < 0.001). Shimmer for vowels [o] and [u] was lower than shimmer for [a] in hoarse patients (p < 0.05). Jitter for [a] was higher than jitter for the other vowels in normal patients. All algorithms improveded with surgical and/or speech therapy. Especially levels of noise for vowels [a] [e] and [o] presented statiscally significant improvements (p < 0.001), shimmer for [a] and [e] (p < 0.001) and jitter for [e] (p < 0.01). The hoarse patients presented values more pathological than the normal subjects, especially jitter for vowels [a] [e] and [o] (p < 0.001), shimmer for [a] [e] [i] and [u] (p < 0.001) and level of noise for [a] [e] [o] and [u] (p < 0.001).

The analysis of vowel [e] presented higher r than [a] when correlated with perceptual evaluation of voice or with voice quality that gave us Dr. Speech 3.0. Also, the [e] is the best vowel to express the improvements of jitter, shimmer and level of noise of hoarse patients after following surgical and/or speech therapy.