Singing and Speaking Voice Characteristics of Trans MtF

Abstract

Compared to other developed countries, in India, the transgender community are often overlooked as a non-existent community. These individuals are considered cursed, punished and therefore are forced to remain completely uninvolved in important matters of the society. Among all of the existing problems already faced by this community, misidentification of their gender based on the voice is a matter of concern. Among this group, most of them are engaged in singing as a part of the daily routine. However, all most all of these trans-singers are not trained in singing. Hence there is a need for this study to analyse characteristics of singing compared to speaking voice of Trans and understand the state of voice to match the age of individual. There were 20 males to female trans-genders with an average age of 35 years (Trans MtF) and 20 females with an average age of 30 years who participated in the study. Normal subjects were ruled out not to be having any kind of speech, language and hearing difficulties. All of them were instructed to inhale deeply, following which they were asked to phonate /a/ in their speaking, followed by singing voice. Three repetitions of phonation were recorded individually in a noise-free environment using a voice recorder; a Sony IC recorder was used for accumulating the voice samples. Praat (Version 5.3.56) software was utilized for the acoustic analysis of the collected data at a sampling frequency of 22 kHz with 12-bit quantization. The parameters considered were speaking fundamental frequency, singing fundamental frequency (SF0) and speaking fundamental intensity and singing fundamental intensity. The long-term average spectrum (LTAS) was analysed separately for 0‑2, 2‑4, 4‑6, 6‑8 and 8‑10 kHz and mean and standard deviation of the same were obtained. Each repetition of phonation were analysed separately, and the average of each parameter was calculated. Obtained data was further statistically analysed using Statistical Package for the Social Sciences (SPSS) software. Mean and standard deviation of each parameter were taken for speaking as well as for singing for both Trans MtF and normal female population. It was noted that the mean of frequency, intensity and LTAS of singing were more than the mean of speaking. Wilcoxon Signed Rank test was administered to see the group on the parameters. Results showed that there was a significant difference between singing and speaking condition for all the parameters in both groups. Mean of parameters were greater in singing than speaking condition. Through the Mann-Whitney U test, results revealed that there was no significant difference between Trans MtF and females except singing fundamental frequency of /a/ (/Z/=2.180, p=0.029) and females had higher mean singing frequency i.e., 294 kHz than Trans MtF (Mean SF0=196 Hz).

Keywords: long-term average spectrum, phonation, singing fundamental frequency, transgender, Trans MtF

Cite this Article

Irfana M., Nissy Sunil, Nutan. Singing and Speaking Voice Characteristics of Trans MtF. Research & Reviews: A Journal of Bioinformatics. 2019; 6(3): 1–4p.