Gender-Affirming Voice Therapy: Why Pitch Is Only Part of the Story
🎯 Key Takeaways
- F0 explains only 41.6% of perceived gender (Leung et al., 2018 meta-analysis)—resonance, intonation, loudness, and articulation account for much of the rest
- Reference F0 bands: female-typical above ~180 Hz, male-typical ~100–140 Hz, with an ambiguous range between—commonly narrowed to ~145–165 Hz
- F0 and formants must shift together—changing either alone rarely changes perceived gender (Hillenbrand & Clark, 2009; Gelfer & Mikos, 2005)
- Testosterone lowers pitch; estrogen does not change the adult voice—and 21% of people on testosterone do not reach male-typical F0 after one year (Ziegler et al., 2018)
- Intonation training now has RCT evidence—randomized trials from the Ghent group show intonation work changes acoustics and listener ratings (Papeleu et al., 2025)
- Client-reported outcomes (TWVQ) are the anchor—acoustic gains and satisfaction do not always align, so measure both
“Just raise your pitch to 200 Hz and you'll sound feminine.” If only it were that simple.
This oversimplification has led to frustration for countless transgender and gender-diverse clients—and the clinicians trying to help them. The research is clear: speaking fundamental frequency explains only 41.6% of the variance in perceived gender (Leung, Oates, & Chan, 2018). The rest comes from resonance, intonation patterns, loudness, articulation, and a constellation of other acoustic and behavioral features.
Two framing points before the numbers. First, voice goals in this area are self-determined: not every trans or gender-diverse person wants voice modification, targets are not always binary, and the clinician's job is to support the client's own goals—not to impose normative ones (WPATH SOC-8; ASHA Practice Portal). Second, acoustic targets are means, not ends: the outcome that matters is how the client experiences their voice in their life.
With that frame, this guide walks through what the evidence says—from perceptual thresholds to formant strategy, hormone effects, therapy delivery formats, and outcome measurement. This September 2026 revision adds the randomized intonation-training trials from Ghent (Papeleu et al., 2025), the large Karolinska–La Trobe acoustic outcome study (Södersten et al., 2024), and corrected, source-checked statistics throughout.
The 41.6% Problem: Why Pitch Alone Isn't Enough
Leung, Oates, and Chan's 2018 systematic review and meta-analysis in the Journal of Speech, Language, and Hearing Research reoriented clinical practice. Across 38 studies, speaking fundamental frequency accounted for 41.6% of the variance in listener gender perception. Pitch, resonance, loudness, articulation, and intonation were all associated with gender perception; tempo and stress were not; findings on breathiness were mixed. The authors also flagged substantial risk of bias in the underlying literature—worth remembering whenever precise-sounding numbers circulate in this field.
The practical reading: more than half of what makes a voice read as masculine or feminine is not pitch. Protocols that focus narrowly on reaching ~200 Hz miss most of the picture.
The Clinical Implication
A client can reach a speaking F0 of 200 Hz and still be misgendered on the phone if resonance and intonation haven't been addressed. Conversely, some speakers with F0 in the ambiguous range are consistently perceived as their target gender because the other dimensions carry the signal. Pitch is one instrument in the ensemble, not the soloist.
What the Meta-Analysis Associates with Gender Perception
Fundamental frequency (41.6% of variance)
The most studied contributor—necessary in most cases, sufficient in few
Resonance / formant frequencies
F1–F3 shaped by vocal tract configuration; the second most studied contributor
Intonation
Pitch variability and contour shapes—now backed by randomized trials
Loudness & articulation
Both associated with perceived gender; tempo and stress were not
Note: the meta-analysis quantified variance only for F0; the remaining contributors are associated with gender perception without validated percentage splits. Beware of precise-looking percentage breakdowns for the “other 58%”—they are not in the source.
Pitch Ranges: The Numbers That Matter
While pitch isn't everything, it remains clinically important, and published reference bands help frame goals (Davies, Papp, & Antoni, 2015; Schwarz et al., 2023):
| Band | Speaking F0 | Typical perception |
|---|---|---|
| Female-typical range | ~180–220 Hz | Usually perceived as feminine—if resonance and prosody align |
| Gender-ambiguous range | ~140–180 Hz (often narrowed to 145–165 Hz) | Ambiguous—other features decide |
| Male-typical range | ~100–140 Hz | Usually perceived as masculine |
What voice training achieves for feminization: in the Schwarz et al. (2023) meta-analysis, speech therapy raised mean F0 by 27 Hz on sustained /a/, 39 Hz in reading, and 25 Hz in spontaneous speech. Total therapy time did not significantly influence F0 gain (p = 0.625)—technique and practice quality matter more than raw session count. In the largest prospective series to date, 74 trans women completing an 8–12-session program showed measurable F0 and formant changes relative to cisgender reference speakers (Södersten et al., 2024).
Sample type matters when you measure. The same meta-analysis found the type of speech sample significantly affected the measured F0 gain (p < 0.01). Reading passages show the largest gains; spontaneous speech the smallest—and spontaneous speech is the ecologically valid one. Collect spontaneous conversation, not just vowels and reading, and compare like-with-like across sessions.
Critical Clinical Point
Acoustic gains and client satisfaction are distinct outcomes that do not always align: across the La Trobe and Ghent program studies, self-ratings, listener ratings, and acoustic measures diverge for a meaningful share of clients—some improve acoustically without feeling better about their voice, and vice versa. Client-reported outcomes (TWVQ, satisfaction) must be measured in their own right, not inferred from hertz.
Formants: The Hidden Key to Gender Perception
Formant frequencies are on average about 20% higher in cisgender women than in cisgender men (as summarized in Schwarz et al., 2023)—a difference reflecting both vocal tract length and learned articulatory behavior. Because part of the difference is behavioral, it is trainable: resonance therapy works by changing vocal tract configuration, and measured formant shifts after training are documented (Carew et al., 2007; Södersten et al., 2024).
F1 (First Formant)
Related to jaw opening and tongue height
Clinical note: contributes to perceived femininity/masculinity alongside the higher formants; raised by greater jaw opening.
F2 (Second Formant)
Related to tongue advancement (front–back)
Clinical note: a central feminization target—raised through anterior tongue carriage and lip spreading.
F3 (Third Formant)
Related to lip configuration and cavity lengths
Clinical note: Carew, Dacakis, and Oates (2007) documented F3 increases and higher listener femininity ratings after a short oral-resonance program (5 sessions).
Landmark Finding: F0 and Formants Travel Together
In perception experiments with resynthesized voices, shifting either F0 or formants alone rarely changed the perceived sex of a voice; shifting both together did so for the large majority of stimuli (Hillenbrand & Clark, 2009; converging with Gelfer & Mikos, 2005, on isolated vowels). This is the core argument for why resonance work must accompany pitch work.
Resonance Modification Techniques
For Voice Feminization (Raising Formants)
- Anterior tongue carriage: Keep tongue positioned more forward in the mouth
- Lip spreading: Spread lips horizontally rather than rounding
- Shortened effective vocal tract: Gentle laryngeal elevation and forward focus
- Oral resonance focus: Direct sound toward the front of the mouth
For Voice Masculinization (Lowering Formants)
- Chest resonance development: Feel vibration in the chest during phonation
- Laryngeal reposturing: Allow larynx to sit in a lower, relaxed position
- Posterior tongue position: Keep tongue slightly retracted
- Lip rounding: Slight lip protrusion elongates the vocal tract
Measurement note: formants should be measured from uploaded recordings, never from live videoconferencing audio—formant estimates from Zoom-transmitted speech deviate substantially from reference recordings, while lossless phone recordings track them well (Zhang et al., 2021). See our formant analysis guide for Praat settings by speaker type, and the telehealth guide for the recording workflow.
Intonation and Prosody: Now with RCT Evidence
Beyond pitch and resonance, prosodic features influence gender perception. The foundational observation is Wolfe, Ratusnik, Smith, and Northrop (1990): among trans women, femininity ratings were associated with more upward intonation contours and fewer level and downward ones. The Ghent group has since carried this from correlation to causation:
Feminine-Associated Prosody
- • Greater pitch variability (wider semitone range)
- • More rising intonation contours
- • Fewer level and falling contours
- • Somewhat softer speaking intensity
- • More precise articulation, larger vowel space
Masculine-Associated Prosody
- • Narrower pitch range within utterances
- • More level and falling contours
- • Stronger loudness
- • Less pitch movement at phrase ends
Randomized Evidence for Intonation Training
Papeleu and colleagues (2025) randomized 30 transfeminine participants to four weeks of intensive intonation training (with a sham-training control phase) and found statistically significant increases in all acoustic intonation parameters, with effects on listener perceptions and self-report examined alongside (JSLHR). A companion randomized trial tested intonation training for people seeking a more masculine-sounding voice (Papeleu et al., AJSLP). Intonation is no longer an afterthought—it is a trainable target with trial evidence of its own.
Hormone Effects: What Changes and What Doesn't
Understanding hormone effects is essential for setting realistic expectations and timing therapy appropriately.
Testosterone (Voice Masculinization)
- • F0 typically descends from female-typical toward male-typical ranges
- • Most change occurs in the first months, with a plateau by around one year
After one year of testosterone, an estimated 21% do not reach cisgender male-typical F0 (≤131 Hz), and 21% report incomplete voice–gender congruence. Voice training is the indicated next step, with pitch-lowering surgery an option in specialist centers.
Estrogen (Voice Feminization)
This is a consistent finding across the peer-reviewed literature. Once vocal folds elongate and thicken during testosterone-driven puberty, feminizing hormones do not reverse those structural changes.
Similarly produce no perceivable voice effects in adults.
→ Voice training and/or surgery are the routes to voice feminization
Evidence-Based Therapy Delivery
ASHA's Practice Portal and the systematic reviews (Leyns et al., 2021; Schwarz et al., 2023) identify a common toolbox for gender-affirming voice work:
Core Therapeutic Techniques
Pitch elevation / lowering training
Systematic work toward the client's target F0 band, generalized from tasks to conversation
Articulation-resonance training
Tongue carriage, lip posture, and oral focus to move formants; combined pitch + resonance programs outperform either alone
Intonation training
Contour and variability work, now supported by randomized trials (Papeleu et al., 2025)
Vocal health foundation
SOVT exercises, easy phonation, and hygiene education to keep modification sustainable and safe
Scheduling Flexibility: Good News for Access
Quinn, Oates, and Dacakis (2024; first published online 2022) compared traditional weekly delivery against an intensive compressed schedule for transfeminine clients and found comparable effectiveness, though their qualitative work notes intensive formats can feel demanding for some participants. Clients can choose scheduling that fits their lives without compromising outcomes—consistent with the meta-analytic finding that total therapy time does not predict F0 gain (Schwarz et al., 2023).
Telepractice works for delivery. Pilot work has delivered intensive feminization programs entirely via telepractice with significant gains in F0 and TWVQ scores, expanding access for clients in underserved areas. Two cautions from our telehealth acoustic guide apply directly here: live-platform audio preserves mean F0—which is exactly what real-time pitch biofeedback needs—but inflates F0 variability on most platforms and corrupts formants, so intonation and resonance measurements belong in uploaded local recordings, not the live stream.
Clinical Assessment: What to Measure and How
ASHA's Practice Portal and WPATH SOC-8 outline assessment domains that go far beyond simple pitch measurement:
| Domain | What to Assess | Measures |
|---|---|---|
| Pitch | Speaking F0 in spontaneous speech, reading, sustained vowels—sample type changes the numbers | Mean F0, F0 range (semitones), F0 SD |
| Resonance | Formant frequencies for corner vowels, from uploaded recordings | F1, F2, F3 values |
| Intonation | Pitch variability, contour patterns | Semitone range, contour classification |
| Voice Quality | Vocal health during modification—strain and fatigue are the risks to catch early | CPPS, jitter, shimmer, HNR; CAPE-V |
| Self-Report | Voice-related quality of life, congruence, satisfaction | TWVQ, TMVQ, VENI |
| Perception | Listener gender attribution, ideally from naïve listeners | VAS masculinity–femininity ratings |
The Trans Woman Voice Questionnaire (TWVQ)
The TWVQ is the most widely used patient-reported outcome for voice feminization. Developed by Dacakis, Davies, Oates, and colleagues (2013)—originally published as the Transsexual Voice Questionnaire (Male-to-Female) and since renamed—it comprises 30 items on a 4-point scale, with total scores from 30 to 120 (higher = greater voice-related difficulty). Validations across languages, including Brazilian Portuguese, Spanish, German, Italian, Turkish, and Chinese, report strong internal consistency and test–retest reliability.
Other Questionnaires
- TMVQ (Trans Man Voice Questionnaire): adaptation of the TWVQ for trans men; validation work is ongoing
- VENI (Voice-related Experiences of Nonbinary Individuals): a validated instrument developed specifically for nonbinary clients, for whom the TWVQ/TMVQ framing may not fit
Surgery and Therapy: A Synergistic Approach
For some clients, voice training alone may not achieve desired outcomes—or they may prefer a combined approach. The Schwarz et al. (2023) meta-analysis quantifies the comparison directly, on the same footing:
| Mean F0 gain (Schwarz et al., 2023) | Sustained /a/ | Reading | Spontaneous speech |
|---|---|---|---|
| Voice therapy | 27 Hz | 39 Hz | 25 Hz |
| Phonosurgery | 72 Hz | 41 Hz | 39 Hz |
Phonosurgery produced significantly larger F0 gains than therapy alone across sample types (p < 0.01), with Wendler glottoplasty—endoscopic suturing of the anterior vocal folds to shorten the vibrating length—as the most common current technique. But F0 is the only thing surgery changes: resonance, intonation, articulation, and prosody remain therapy's territory, which is the case for combining the two.
The Case for Combined Intervention
Post-operative voice therapy addresses everything surgery cannot, and case series report better combined acoustic, perceptual, and self-reported outcomes when structured therapy follows glottoplasty than with surgery alone (e.g., Casado et al., 2017). Pre-operative therapy lets clients test how far behavioral change takes them before committing to an irreversible procedure.
For transmasculine clients whose F0 remains above target on testosterone (the 21% in Ziegler et al., 2018), pitch-lowering surgery (relaxation/Type III thyroplasty) is an option in specialist centers, with voice training the first-line and complementary approach.
WPATH and ASHA: Clinical Frameworks
WPATH Standards of Care Version 8 (2022)
- • Includes a dedicated Voice and Communication chapter
- • Developed by an international expert panel using Delphi methodology
- • Recommends health care providers receive education on vocal functioning in trans health
- • Cites evidence that voice training effectively increases F0, satisfaction, and quality of life
ASHA Practice Portal
- • Confirms gender-affirming voice services are within SLP scope of practice
- • Emphasizes person-centered care with client-determined goals
- • Recommends avoiding binary constructs when they don't fit the client
- • Coverage and reimbursement vary by payer and jurisdiction—and have shifted repeatedly since 2024–2025, so verify current policy rather than relying on published summaries
The Evidence Base: Improving, Still Limited
Leung et al. (2018) flagged substantial risk of bias across the perception literature, and the treatment literature has long leaned on small, uncontrolled studies. This is changing: randomized trials (Papeleu et al., 2025) and larger prospective cohorts (Södersten et al., 2024; N = 74) are arriving. Still, long-term follow-up and standardized protocols remain scarce—hold clinical certainty proportional to the evidence, and keep client-reported outcomes at the center.
Common Questions
Q: How long does gender-affirming voice therapy typically take?
Programs of 8–12 sessions are typical (that was the format in Södersten et al.'s 74-participant study), plus consistent home practice. Notably, the Schwarz et al. (2023) meta-analysis found total therapy time did not significantly predict F0 outcomes (p = 0.625)—technique quality and practice consistency matter more than raw session count, and delivery format (weekly vs. intensive) can follow the client's life (Quinn et al., 2024).
Q: Can I track progress with PhonaLab?
Yes. Our free Pitch Visualizer provides real-time F0 tracking plus F1, F2, and F3 formant analysis—the parameters that matter for gender-affirming voice work. Use the live view for in-session pitch biofeedback, and uploaded recordings for formant and intonation measurement across sessions. Audio is processed in memory and never stored, which matters to many clients in this population for whom privacy is a safety consideration.
Q: Are smartphone recordings valid for tracking progress?
Yes, for the parameters this work needs. F0 is the most robust measure across recording chains, and phone-recorded CPP correlates at r ≈ .99 with research-grade recording when a headset and quiet room are used (Awan et al., 2024). Formants track well from lossless phone recordings but not from live videoconferencing (Zhang et al., 2021). Keep the same device, distance, and environment across sessions, and see our smartphone recording guide for the full protocol.
Q: What about nonbinary voice goals?
Start from the client's goals, not a binary template. Work collaboratively to define a communication profile from combinations of pitch, resonance, intonation, and articulation features. The gender-ambiguous F0 band can be a deliberate target for some clients, but it is one option among many—resonance and prosody choices allow wide personalization. The VENI questionnaire was developed specifically for tracking outcomes in this population.
Q: What vocal health risks should I monitor?
Modifying voice without proper guidance can cause harm. Muscle tension dysphonia, vocal fatigue, and phonotrauma are risks when clients push toward targets with effortful technique or attempt self-taught modification. Build vocal health foundations into every program, monitor voice quality acoustically and perceptually, and refer for laryngoscopy if a client reports pain, persistent strain, or voice loss.
Q: How should I handle clients whose self-perception doesn't match acoustic progress?
This is common and well documented. Across the La Trobe and Ghent program studies, self-ratings, listener ratings, and acoustic measures do not always align—clients can improve acoustically without feeling better about their voice, and vice versa. Regular perceptual feedback from naïve listeners and systematic use of validated questionnaires (TWVQ) help bridge the gap, and the client's own experience remains the primary outcome.
Bottom Line: Evidence-Based Takeaways
- 1F0 explains 41.6% of perceived gender—resonance, intonation, loudness, and articulation carry the rest (Leung et al., 2018)
- 2Reference bands, not rigid targets: female-typical above ~180 Hz, male-typical ~100–140 Hz, ambiguous between—and goals are the client's to set
- 3F0 and formants must move together for reliable perceptual change—resonance work is not optional (Hillenbrand & Clark, 2009)
- 4Testosterone lowers pitch but 21% don't reach male-typical F0 in a year; estrogen doesn't change the adult voice—training (and sometimes surgery) fills the gap (Ziegler et al., 2018)
- 5Intonation training has randomized-trial support—give it dedicated attention (Papeleu et al., 2025)
- 6Acoustics and satisfaction diverge—client-reported outcomes (TWVQ) are the anchor, measured in their own right
- 7Surgery and therapy are complementary—surgery moves F0 further; therapy owns everything else (Schwarz et al., 2023)
- 8Measure with the right chain: spontaneous speech samples, live F0 biofeedback is fine, but formants and intonation from uploaded recordings only
📊 Track Voice Feminization/Masculinization Progress
Our free Pitch Visualizer provides real-time F0 tracking with F1, F2, and F3 formant analysis—exactly the parameters you need for gender-affirming voice work. Visualize progress across sessions and generate reports showing movement toward client-defined target ranges. Audio is processed in memory and never stored.
Try Free Pitch Visualizer →Real-time pitch tracking • F1-F3 formant analysis • Target range visualization • No installation required
⚠️ Clinical Documentation Tool
The information in this article is provided for educational purposes and clinical workflow support. Voice services should be provided by qualified speech-language pathologists with training in gender-affirming care. Acoustic measures should be interpreted alongside client-reported outcomes and clinical judgment. All treatment decisions should prioritize client safety, autonomy, and self-determined goals. Reimbursement and regulatory conditions vary by jurisdiction and change over time. PhonaLab tools support clinical documentation but do not provide medical diagnoses.
References & Further Reading
- Leung Y, Oates J, Chan SP. (2018). Voice, articulation, and prosody contribute to listener perceptions of speaker gender: A systematic review and meta-analysis. Journal of Speech, Language, and Hearing Research, 61(2), 266–297. doi:10.1044/2017_JSLHR-S-17-0067
- Coleman E, Radix AE, Bouman WP, et al. (2022). Standards of Care for the Health of Transgender and Gender Diverse People, Version 8. International Journal of Transgender Health, 23(sup1), S1–S259. doi:10.1080/26895269.2022.2100644
- Schwarz K, Cielo CA, Spritzer PM, Villas-Boas AP, Costa AB, Fontanari AMV, Costa Gomes B, da Silva DC, Schneider MA, Lobato MIR. (2023). Speech therapy for transgender women: An updated systematic review and meta-analysis. Systematic Reviews, 12(1), 128. doi:10.1186/s13643-023-02267-5
- Hillenbrand JM, Clark MJ. (2009). The role of f0 and formant frequencies in distinguishing the voices of men and women. Attention, Perception, & Psychophysics, 71(5), 1150–1166.
- Gelfer MP, Mikos VA. (2005). The relative contributions of speaking fundamental frequency and formant frequencies to gender identification based on isolated vowels. Journal of Voice, 19(4), 544–554. doi:10.1016/j.jvoice.2004.10.006
- Davies S, Papp VG, Antoni C. (2015). Voice and communication change for gender nonconforming individuals: Giving voice to the person inside. International Journal of Transgenderism, 16(3), 117–159.
- Wolfe VI, Ratusnik DL, Smith FH, Northrop G. (1990). Intonation and fundamental frequency in male-to-female transsexuals. Journal of Speech and Hearing Disorders, 55(1), 43–50. doi:10.1044/jshd.5501.43
- Carew L, Dacakis G, Oates J. (2007). The effectiveness of oral resonance therapy on the perception of femininity of voice in male-to-female transsexuals. Journal of Voice, 21(5), 591–603. doi:10.1016/j.jvoice.2006.05.005
- Dacakis G, Davies S, Oates JM, Douglas JM, Johnston JR. (2013). Development and preliminary evaluation of the Transsexual Voice Questionnaire for Male-to-Female Transsexuals [now the Trans Woman Voice Questionnaire]. Journal of Voice, 27(3), 312–320.
- Ziegler A, Henke T, Wiedrick J, Helou LB. (2018). Effectiveness of testosterone therapy for masculinizing voice in transgender patients: A meta-analytic review. International Journal of Transgenderism, 19(1), 25–45. doi:10.1080/15532739.2017.1411857
- Papeleu T, Oates J, Tomassen P, Adriaansen A, Leyns C, Alighieri C, Krenn H, D'haeseleer E. (2025). Effects of intensive intonation training in transgender and gender diverse people aiming for a more feminine-sounding voice: A randomized clinical trial. Journal of Speech, Language, and Hearing Research, 68(4), 1711–1742. doi:10.1044/2024_JSLHR-24-00563
- Papeleu T, Oates J, Tomassen P, T'Sjoen G, Leyns C, Daelman J, D'haeseleer E. (2026). Effects of intensive intonation training in transgender and gender diverse people aiming for a more masculine-sounding voice: A randomized clinical trial. American Journal of Speech-Language Pathology, 35(2), 515–545. doi:10.1044/2025_AJSLP-25-00238
- Quinn S, Oates J, Dacakis G. (2024). The effectiveness of gender affirming voice training for transfeminine clients: A comparison of traditional versus intensive delivery schedules. Journal of Voice, 38(5), 1250.e25–1250.e52. doi:10.1016/j.jvoice.2022.03.001
- Södersten M, Oates J, Sand A, Granqvist S, Quinn S, Dacakis G, Nygren U. (2024). Gender-affirming voice training for trans women: Acoustic outcomes and their associations with listener perceptions related to gender. Journal of Voice. Advance online publication.
- Leyns C, Papeleu T, Tomassen P, T'Sjoen G, D'haeseleer E. (2021). Effects of speech therapy for transgender women: A systematic review. International Journal of Transgender Health, 22(4), 360–380. doi:10.1080/26895269.2021.1915224
- Zhang C, Jepson K, Lohfink G, Arvaniti A. (2021). Comparing acoustic analyses of speech data collected remotely. The Journal of the Acoustical Society of America, 149(6), 3910–3916. doi:10.1121/10.0005132
- Awan SN, Bahr R, Watts S, Boyer M, Budinsky R, Bridge2AI Voice Consortium, Bensoussan Y. (2024). Validity of acoustic measures obtained using various recording methods including smartphones with and without headset microphones. Journal of Speech, Language, and Hearing Research, 67(6), 1712–1730. doi:10.1044/2024_JSLHR-23-00759
- ASHA Practice Portal: Gender Diverse Voice and Communication