Acoustic and Perceptual Effects of Intonation Training in Gender Diverse People

The aim of this project is to investigate the short- and the longer-term effect of intonation training on vocal characteristics, listener perceptions and patient related outcome measures (PROMS) in gender diverse people using a randomized sham-controlled trial.

Background and Problem Statement. Speech training in gender diverse individuals focuses on aspects of communication that strongly influence gender perception. Findings from the literature suggest that a listener's gender perception is determined not only by pitch, but also by other components such as intonation. However, the nature and extent of pitch changes required to alter gender perception are not known. In speech therapy, intonation training is frequently used for speech feminization. However, empirical evidence for this is lacking. Research questions and/or hypotheses Thisresearch investigates the short- and long-term effects of intonation training on acoustic voice characteristics, gender perception and findings of gender-diverse individuals using a randomized sham-controlled trial (RCT). Based on findings from the literature, (a) gender-congruent values are expected for intonation parameters, in addition (b) assessments of gender that fit the gender identity of the person and (c) an improvement in self-assessment after intonation training in the short and long term. This study will lead to new insights into the relationship between intonation and gender perception and makes an important contribution to the evidence for the use of intonation training. Research Protocol An RCT will be used. Gender diverse individuals will be recruited through the Ghent University Hospital. Inclusion criteria are: diagnosis of gender dysphoria established by the multidisciplinary gender team, age between 18 and 60 years, normal hearing and Dutch mother tongue. Taking into account the expertise of the research group and the low-threshold recruitment through the gender clinic of UZ Gent, a total of n=20 participants is a very feasible number. Subjects will be randomly assigned to the experimental (intonation training) or control group (receiving a placebo intervention). Randomization will be stratified by age group and baseline speaking fundamental frequency. The experimental group will receive 4 weeks (1h/week) of intonation training with a pre- and post-voice assessment and a follow-up assessment after 8 weeks. The control group will receive sham therapy at a similar duration and frequency, also with a pre- and post-measurement and a follow-up assessment after 8 weeks. The control group will receive 4 more weeks (1 hour/week) of intonation training with a post measurement (due to ethical reasons). The multidimensional speech assessments will be blinded and completed and well-defined training programs will be used. The primary outcome measures are: acoustic intonation parameters, gender perception by listeners and self-assessment and client satisfaction. Linear mixed models with post-hoc pairwise comparisons will be used to analyze the evolution of continuous data within and between groups. ICC models and Cohen's kappa will be used to analyze inter- and intra rater reliability.

Speech training: 4 times 1 hour of intonation training for 4 weeks Standard voice assessment (voice recording, VAS, questionnaire(s)) (1 premeasurement and 2 postmeasurements)

Speech training: : 4 times 1 hour sham therapy (info sessions + active intervention non-verbal communication) during 4 weeks + 4 times 1 hour intonation training during 4 weeks Standard voice assessment (voice recording, VAS, questionnaire(s)) (1 premeasurement and 3 postmeasurements)

The participants will receive intonation training (1h/week) for four weeks, aiming for more gender congruent intonation patterns.

The participants will first receive four weeks of sham therapy (with no expected effect on speech), during which information on voice functioning, voice surgery, voice hygiene, and transgender health care will be provided. In addition, during sham therapy, active intervention will be given around nonverbal communication. After four weeks of sham therapy, for ethical reasons, subjects in this group will also receive four weeks of intonation training (aiming for more gender congruent intonation patterns).

The general intonation shift is the mean size of the general increase or decrease of the fundamental frequency during the utterance. For the calculation of this parameter, the first and last vowel of the sentence were selected and the fo of the vowels were determined. Subsequently, the general intonation shift was calculated as the difference between start and end fo. Based on the size of this difference, the general intonation shift was categorized as an upward, downward or flat intonation shift. A difference smaller than 1.570884399 ERB was considered as a flat intonation shift. Finally, the percentage of upward, downward and flat intonation shifts could be determined.

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 20 min.)

The Final intonation shift indicates the average size of the upward or downward intonation shift at the end of the utterance. This parameter was calculated as the difference between the fo at the beginning of the final rise or drop and the fo at the end. The beginning was located on the last stressed vowel and the end of the final rise or drop was located on the last unstressed vowel. In case that the utterance ended with a stressed vowel, the intonation shift was derived from the difference in fo within this vowel.

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 20 min.)

This parameter was calculated as the difference between the minimum (5th percentile) and the maximum fo (95th percentile) of the utterance. For this calculation the "Intonantion and stress trajectory" script was used (Corthals, 2019a; Van Borsel, Vandaele, & Corthals, 2013).

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 20 min.)

This parameter was determined by the sum of the absolute value of all pitch changes during the utterance between percentile 5 and 95 divided by the total duration of the utterance. The pitch variation index reflects the size and speed of the intonation maneuvers. For this parameter the script "Intonation and stress trajectory" was used (Corthals, 2019a; Van Borsel, Vandaele, & Corthals, 2013).

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 20 min.)

A listening experiment with femininity ratings using a visual analogue scale (VAS: very feminine - neutral - very masculine).

Visual analoge scales investigating the client's satisfaction of the therapy and psychosocial functioning: examples: My speech fits who I am (don't agree at all - neutral - completely agree) How masculine/feminine do you sound while speaking? (very masculine - gender ambiguous (neutral) - very feminine) While speaking, there is a wide variation of my pitch. (don't agree at all - neutral - completely agree) (3) I sound more feminine/male/neutral compared to before the training period. (don't agree at all - neutral - completely agree) After the training period, my pitch range increased (I can produce higher and/or lower tones). (don't agree at all - neutral - completely agree))

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

Trans Women Voice Questionnaire: 30 questions investigating trans women's experiences with their voice (1 = never or seldom; 2 = sometimes; 3 = often; 4 = usual or always). The score can range from 30 to 120. A lower score indicates higher satisfaction with the voice.

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

Voice Handicap Index: 30 questions investigating the impact of voice on the quality of life (0 = never; 1 = almost never; 2 = sometimes; 3 = almost always; 4 = always) . The score can range from 30 to 120. A lower score indicates higher satisfaction with the voice.

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

baseline (before treatment) - at the end of the treatment period (after 4 weeks) - 4 (and 8) weeks after the end of the treatment period (after 8 (and 12) weeks (each measurement +/- 5 min.)

Inclusion Criteria: Normal hearing (using ISO standard audiometry) Native Dutch speaker Exclusion Criteria: A history of neurological disorders Previous phonosurgery or speech therapy Vocal pathology (using videolaryngostroboscopy) Smokers.