Funding was terminated by NIH in 2025 with a statement that the research no longer meets agency priorities.

The goal of this clinical trial is to learn if prototype software can be used to train vocal pitch and resonance in trans women. The main questions it aims to answer are:

• How regularly would trans women use the prototype voice training software as "homework" between human-led voice training sessions?
• In such a setting, how usable and motivating would the training software be?

Participants will first have their voice measured, then take part in four virtual sessions (one per week) led by a speech-language pathologist. In each session, the clinician will guide the participant through several voice exercises both with and without the software. Participants will be asked to use the software to exercise on their own for about 2x 15 minutes a day between virtual sessions.

In an initial virtual assessment session (conducted via a password-protected Zoom room), a researcher will explain the protocol, obtain informed consent, collect participant demographics and other self-report data, and obtain baseline recordings of the participant's voice while speaking different utterances.

After the initial assessment session, each participant will take part in four virtual training sessions (on day 1, after approximately 1 week, after approximately 2 weeks, after approximately 3 weeks) provided on an individual basis by a certified speech-language pathologist (SLP).

In the first session (1 hour), the SLP will assess each participant's baseline ability to modify their pitch and resonance and will introduce the voice training software. They will provide strategies that participants may use to manipulate their pitch or resonance (e.g., for resonance, changing the positioning of the tongue in their mouth or changing the shape of the lips). For the remainder of the session, the SLP will guide the participant through exercises deemed appropriate to the participant's ability, both with and without the software. At the end of the session, the SLP will prescribe pitch and/or resonance exercises to be performed at home with the voice training software, with a suggested duration of 2x 15 minutes each day. Additional one-hour virtual training sessions with the SLP will be held at the end of weeks 1, 2, and 3.

Between sessions, participants will be asked to exercise on their own and report the duration and type of practice using a self-report log, though there will be no mechanism to enforce the unsupervised exercise.

After the last training session, participants will take part in a final assessment session where the same recordings as in the initial assessment session will be obtained again and additional self-report data will be collected.

Inclusion Criteria:

• At least 18 years old.
• Native speaker of American English - must have been exposed to it by the age of 6.
• Assigned male at birth, currently identifies as transgender.
• Currently interested in gender-affirming voice training.
• Has access to a computer and quiet room for performing voice training.
• Able to pass a pure-tone hearing screening at 20dB HL.

Exclusion Criteria:

• Current or previous diagnosed speech, language or hearing problems (besides voice-gender dysphoria).
• Neurological disorder or disease (e.g., Parkinson's, tremor) or impacted nerve function to the larynx (e.g., paralysis/paresis of the vocal fold).
• Respiratory disease (e.g., asthma).
• Nasopharyngeal pathology or anatomical abnormality.
• Previous voice feminization surgery.
• Engaged in other gender-affirming voice training during period of study participation.

transgender women and nonbinary transfeminine participants

We plan to share all anonymized data with other researchers. This is expected to include questionnaire answers and acoustic metrics extracted from data, but not, e.g., audio recordings or specific participant data that would allow the participant to be individually identified (e.g., date of birth).