Find a quiet room and hold your phone about a hand’s width from your mouth.

Leave a clear two-second pause between the parts so each one can be found.

On iPhone, record in Voice Memos, then tap Share and Save to Files so you can pick it here.

Each row shows the earlier check first and the later one second. Where your goal has a target range, the change is marked as closer to or further from it.

Small differences can come from the room, the microphone or the day, so look for changes that hold across several checks. Recording in the same quiet place each time makes checks easier to compare.

Pitch is how high or low your voice sounds. It’s measured in hertz (Hz): how many times your vocal folds vibrate each second.

Typical speaking pitch is about 90 to 155 Hz for men and 165 to 255 Hz for women. Between 145 and 165 Hz is described as an overlap zone, where listeners could hear either.

Resonance is the size of the space your voice rings in above the vocal folds. A smaller space sounds brighter; a larger one sounds darker.

Hum measures it from the first three formants of your held “ee” and “ah”, then places you on a scale where 0 is the typical men’s average and 100 the typical women’s.

Listeners rely on resonance heavily: a 2021 study of 379 speakers found it mattered more than earlier research suggested, and another found some evidence that listeners used F2 alongside pitch when rating femininity (Houle and Levi 2021).

Hum also estimates resonance from all the vowels in your reading, which is closer to what people hear day to day. It uses a fixed analysis setting, so raising your pitch doesn’t move it on its own: in tests, synthetic voices with a typical men’s vocal tract stayed near 0 whether they spoke at 115 or 210 Hz.

Your range is the span from the lowest to the highest comfortable note in your siren. It’s shown in hertz and in semitones; 12 semitones make an octave.

Melody, or intonation, is how your pitch moves through a phrase: rising, falling or staying level. Hum measures how far your pitch travels each second, how wide your phrases are, and whether each phrase ends rising, falling or level.

In a 2025 study that included trans speakers, more pitch movement and fewer flat phrases went with higher femininity ratings, and flatter phrases with more masculine ratings. Earlier studies had mixed results, so treat melody as one ingredient among several (Papeleu et al. 2025).

Vocal weight is how light and soft or full and buzzy your voice sounds, separate from how high or low it is. Hum measures it as H1–H2: how much stronger the first harmonic of your voice is than the second, in decibels. When the vocal folds close less firmly the lowest harmonic stands out more and the number rises, and firmer closure lowers it (Hanson 1997). On the held “ah” it is corrected for the vowel’s formants (H1*–H2*; Iseli and Alwan 2004), and adults have averaged about 1.5 dB for men and 3.5 dB for women, with plenty of overlap (Iseli, Shue and Alwan 2006). It’s a smaller cue to gender than pitch and resonance (Södersten et al. 2024), and loudness, pitch, microphones and the room all move it, so watch your own trend in checks made the same way.

The Light and full exercise works on it the same way, but compares each hold with your own usual “ah” at the same pitch and loudness, so microphones and rooms don’t get in the way.

Pace is how many syllables you say per second. Hum finds syllables from peaks in loudness, following de Jong and Wempe (2009), and each reading passage has a known syllable count (the first one has 60). Pauses are silences of at least 0.3 seconds.

Pace isn’t a dependable gender cue: a study of audiobook narrators found no difference between men and women (Stolarski 2019). It’s shown so you can see how you read.

Steadiness is how much your pitch wobbles while you hold a vowel, in cents; 100 cents make a semitone. Lower is steadier.

Clarity is the harmonics-to-noise ratio (HNR): how much of the sound is clear tone rather than breath or noise. A reference sheet from the Sydney Voice Lab lists 20 dB and above as typical of voices without a disorder (Warhurst et al. 2012). Phone microphones and room noise pull it down, so use it as a rough guide.

The same sheet notes that adults typically hold an “ah” for about 15 to 20 seconds (Goy et al. 2013), which the Long “ah” exercise measures.

Keep sessions short and easy, and stop if your throat hurts or your voice goes hoarse. Warm up first and finish with an easy hum.

ASHA, the US professional body for speech and language therapists, warns that changing your voice without guidance can lead to strain, and that pitch-tuning apps aren’t designed for voice work and can encourage strained practice. If you can, use tools like this alongside a therapist.

The NHS advises seeing a GP if hoarseness hasn’t improved after 2 weeks.

A speech and language therapist who works in gender-affirming voice care can assess you in person and tailor the work to your goals.

In Ireland, the HSE National Gender Service has speech and language therapists on its team.

ASHA suggests practising in realistic situations, such as phone and video calls or ordering in a café, so a new voice carries over into everyday life.

The Hum into words and Everyday phrases exercises are a step toward that: they start from a sound where your target is easy to find and move it into real speech.

Conversation practice goes further: short role-plays, like ordering a coffee or taking a call from a friend, where Hum plays the other person and scores how much of each reply stays in your zone. The voice monitor listens while you talk freely, and chimes softly if you drift out of your range for a few seconds.

• Straw slides are a semi-occluded vocal tract exercise: partly closing the mouth changes the air pressure above the vocal folds in a way that helps them vibrate easily (Titze 2006). ASHA lists them for gender-affirming voice work.
• Easy hum comes from resonant voice therapy, which a systematic review found has a moderate level of evidence for easier, clearer voicing (Yiu et al. 2017).
• Pitch ladder moves pitch in small steps. ASHA describes using vocal function exercises to safely shift the pitch range, and a systematic review found those exercises effective across normal and disordered voices (Angadi et al. 2019).
• Resonance meter and Dark and bright practise vowel modification and formant shifting, which ASHA lists as resonance techniques.
• Melody match practises the intonation patterns described under What is melody.
• Long “ah” measures maximum phonation time, a standard voice check (Goy et al. 2013).
• Light and full works on vocal weight, measured live as described under What is vocal weight.
• Conversation practice and the voice monitor carry your voice into realistic speech, as ASHA recommends.
• Programs put these together into short sessions that unlock week by week, from warm-ups to everyday speech.

Live scoring uses the same measurements as the voice check, updated several times a second, so treat it as a guide rather than a verdict.

Pitch is measured with the YIN method. Resonance comes from the first three formants of your held vowels, compared with the averages of 45 men and 48 women. Tested on those recordings, Hum put 89 of the 93 speakers on the expected side of the resonance scale.

Resonance in speech averages the formants of every vowel in your reading at a fixed 5000 Hz analysis setting; on the same speakers it put 91 of 93 on the expected side. Background noise and clipping push it toward the middle, so it’s only shown when the recording is clean enough.

Melody comes from the pitch track, split into phrases at pauses of a quarter second. Pace follows de Jong and Wempe (2009). Clarity uses the harmonics-to-noise method of Boersma (1993), the one Praat uses, and Hum’s values were checked against Praat’s. Siren breaks are sudden steps of 3 semitones or more, or dropouts, in the middle of the slide.

Accents shift some vowels, so your own trend over time matters more than any single score. Reading passages come in seven languages, but the typical ranges and resonance averages come mostly from studies of American English speakers: in another language, compare yourself with your own earlier checks rather than with the averages.

Your audio is analysed on this device and never uploaded. With the voice diary on, the reading of each check you save is kept on this device too, so you can listen back; it’s never synced or put in backups. This is a practice aid, not a medical assessment.

Every source is a peer-reviewed paper, a clinical textbook or a professional body. Links go to the original publication.

1. Baken RJ, Orlikoff RF. Clinical Measurement of Speech and Voice, 2nd edition. Singular, 2000. Typical speaking pitch for men and women.
2. Speech therapy for transgender women: an updated systematic review and meta-analysis. Systematic Reviews, 2023. The 145 to 165 Hz overlap zone, and average pitch gains after therapy and after surgery. doi:10.1186/s13643-023-02267-5
3. Gelfer MP, Schofield KJ. Journal of Voice, 2000: trans women heard as women spoke at about 164 to 199 Hz. As cited in Acoustic measures of Brazilian transgender women’s voices, Frontiers in Psychology, 2021. doi:10.3389/fpsyg.2021.622526
4. American Speech-Language-Hearing Association. Gender-affirming voice and communication, Practice Portal. Individual and gender-neutral goals, about 145 to 175 Hz as an androgynous pitch, techniques, what testosterone changes, and cautions about practising without guidance. asha.org
5. Leung Y, Oates J, Chan SP. 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, 2018. doi:10.1044/2017_JSLHR-S-17-0067
6. Leung Y, Oates J, Chan SP, Papp V. Associations between speaking fundamental frequency, vowel formant frequencies, and listener perceptions of speaker gender and vocal femininity-masculinity. Journal of Speech, Language, and Hearing Research, 2021. doi:10.1044/2021_JSLHR-20-00747
7. Hillenbrand J, Getty LA, Clark MJ, Wheeler K. Acoustic characteristics of American English vowels. Journal of the Acoustical Society of America 97(5):3099-3111, 1995. doi:10.1121/1.411872
8. Glottoplasty for vocal feminization in transgender women: position paper of the Brazilian Academy of Laryngology and Voice. Brazilian Journal of Otorhinolaryngology. scielo.br
9. Effect of Wendler glottoplasty on voice and quality of life of transgender women. Brazilian Journal of Otorhinolaryngology, 2023. Hormone treatment has no influence on the voice of trans women. PMC9874341
10. Hodges-Simeon CR, Grail GPO, Albert G, Groll MD, Stepp CE, Carré JM, Arnocky SA. Testosterone therapy masculinizes speech and gender presentation in transgender men. Scientific Reports 11:3494, 2021. doi:10.1038/s41598-021-82134-2
11. Vendramini BDV, Campos LMS, de Moraes BT, de Vasconcelos SJ, Aires MM. Type III thyroplasty for patients with high-pitched voice disorders: a systematic review and meta-analysis. Otolaryngology-Head and Neck Surgery 172(6):1847-1855, 2025. doi:10.1002/ohn.1185
12. de Cheveigné A, Kawahara H. YIN, a fundamental frequency estimator for speech and music. Journal of the Acoustical Society of America 111(4):1917-1930, 2002. doi:10.1121/1.1458024
13. Irish Association of Speech and Language Therapists, Equality special interest group: speech and language therapy at the HSE National Gender Service. iaslt.ie
14. Papeleu T, Leyns C, Tomassen P, T’Sjoen G, Van Lierde KM, D’haeseleer E. Intonation parameters in gender diverse people. Journal of Voice 39(3):845.e23-845.e38, 2025. Also summarises Wolfe et al. (1990) and Hancock et al. (2014). doi:10.1016/j.jvoice.2022.12.020
15. Houle N, Levi SV. Effect of phonation on perception of femininity/masculinity in transgender and cisgender speakers. Journal of Voice 35(3):497.e23-497.e37, 2021. doi:10.1016/j.jvoice.2019.10.011
16. Titze IR. Voice training and therapy with a semi-occluded vocal tract: rationale and scientific underpinnings. Journal of Speech, Language, and Hearing Research 49(2):448-459, 2006. doi:10.1044/1092-4388(2006/035)
17. Yiu EML, Lo MCM, Barrett EA. A systematic review of resonant voice therapy. International Journal of Speech-Language Pathology 19(1):17-29, 2017. doi:10.1080/17549507.2016.1226953
18. Angadi V, Croake D, et al. Effects of vocal function exercises: a systematic review. Journal of Voice 33(1):124.e13-124.e34, 2019. ASHA Evidence Map summary
19. de Jong NH, Wempe T. Praat script to detect syllable nuclei and measure speech rate automatically. Behavior Research Methods 41(2):385-390, 2009. doi:10.3758/BRM.41.2.385
20. Stolarski Ł. Lack of effects of gender on the reading rate of long texts. Sociolinguistic Studies 12(3-4):461-479, 2019. journal.equinoxpub.com
21. Boersma P. Accurate short-term analysis of the fundamental frequency and the harmonics-to-noise ratio of a sampled sound. Proceedings of the Institute of Phonetic Sciences 17:97-110, 1993. fon.hum.uva.nl
22. Sydney Voice Lab, University of Sydney. Acoustic analysis normative data reference sheet, April 2025. Typical maximum phonation time (Goy et al. 2013), harmonics-to-noise reference (Warhurst et al. 2012) and average speaking pitch (Leung et al. 2022). sydney.edu.au
23. NHS. Laryngitis: when to see a GP. nhs.uk
24. Hanson HM (1997). Glottal characteristics of female speakers: Acoustic correlates. Journal of the Acoustical Society of America 101(1):466–481. A more open glottis gives relatively stronger low and weaker high harmonics; H1*–H2* as a measure of it. doi:10.1121/1.417991
25. Hanson HM, Chuang ES (1999). Glottal characteristics of male speakers: Acoustic correlates and comparison with female data. Journal of the Acoustical Society of America 106(2):1064–1077. H1*–H2* averaged 0.0 dB in 21 men (SD 1.8) and 3.1 dB in 22 women (SD 2.0); repeated tokens varied by about 1 dB within a speaker. doi:10.1121/1.427116
26. Klatt DH, Klatt LC (1990). Analysis, synthesis, and perception of voice quality variations among female and male talkers. Journal of the Acoustical Society of America 87(2):820–857. Uncorrected H1–H2 about 5.7 dB higher in women than in men (as summarised by Hanson and Chuang 1999). doi:10.1121/1.398894
27. Iseli M, Alwan A (2004). An improved correction formula for the estimation of harmonic magnitudes and its application to open quotient estimation. Proceedings of IEEE ICASSP 2004, vol. 1, pp. I-669–I-672. The formula that removes the formants’ boost from the harmonic levels (H1*–H2*). doi:10.1109/ICASSP.2004.1326074
28. Iseli M, Shue Y-L, Alwan A (2006). Age- and gender-dependent analysis of voice source characteristics. Proceedings of IEEE ICASSP 2006, vol. 1, pp. I-389–I-392. Adults aged 20 to 39 averaged H1*–H2* of about 1.5 dB (men) and 3.5 dB (women), with much overlap: the zone edges. doi:10.1109/ICASSP.2006.1660039
29. Iseli M, Shue Y-L, Alwan A (2007). Age, sex, and vowel dependencies of acoustic measures related to the voice source. Journal of the Acoustical Society of America 121(4):2283–2295. Women about 3.4 dB above men; in low voices H1*–H2* rises with pitch (about 0.22 dB per Hz, 80 to 175 Hz); /i/ about 3 dB below /ɪ/. doi:10.1121/1.2697522
30. Hawks JW, Miller JD (1995). A formant bandwidth estimation procedure for vowel synthesis. Journal of the Acoustical Society of America 97(2):1343–1344. The formant bandwidths used in the correction, as in VoiceSauce. doi:10.1121/1.412986
31. Shue Y-L, Keating P, Vicenik C, Yu K (2011). VoiceSauce: A program for voice analysis. Proceedings of the 17th International Congress of Phonetic Sciences, Hong Kong, pp. 1846–1849. How harmonic levels are found and corrected (pitch-synchronous windows, a peak search around each harmonic, F1 and F2 corrections). www.internationalphoneticassociation.org
32. Södersten M, Lindestad P-Å (1990). Glottal closure and perceived breathiness during phonation in normally speaking subjects. Journal of Speech and Hearing Research 33(3):601–611. Softer phonation closed the glottis less completely and sounded breathier; pitch had no significant effect. Loudness moves weight. doi:10.1044/jshr.3303.601
33. Van Borsel J, Janssens J, De Bodt M (2009). Breathiness as a feminine voice characteristic: a perceptual approach. Journal of Voice 23(3):291–294. For all seven women recorded, a breathy /a/ was rated more feminine than their usual one. doi:10.1016/j.jvoice.2007.08.002
34. Zhang Z, Zhang J, Kreiman J (2022). Effects of laryngeal manipulations on voice gender perception. Proceedings of Interspeech 2022, pp. 1856–1860. In synthetic voices F0 had a large effect on perceived gender and spectral shape a moderate one; a low F0 with a high H1–H2 added ambiguity. doi:10.21437/Interspeech.2022-10815
35. Nylén F, Holmberg J, Södersten M (2024). Acoustic cues to femininity and masculinity in spontaneous speech. Journal of the Acoustical Society of America 155(5):3090–3100. Across 132 voices, median pitch was the primary cue to perceived femininity and masculinity. doi:10.1121/10.0025932
36. 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 40(4):1242.e31–1242.e59 (2026 issue; online 2024). In 74 trans women, pitch, formants and loudness best predicted a female-sounding voice; some raised L1–L2 (H1–H2) with training. doi:10.1016/j.jvoice.2024.02.003
37. Penney J, Gibson A, Cox F, Proctor M, Szakay A (2021). A comparison of acoustic correlates of voice quality across different recording devices: a cautionary tale. Proceedings of Interspeech 2021, pp. 1389–1393. Phones and laptops changed voice-quality measures such as H1–H2, by amounts that depended on the value. doi:10.21437/Interspeech.2021-729