Written by Lucas Reed | Last updated: June 2026
About the Tool
What does the Deep Voice Test measure? The tool measures the fundamental frequency (F0) of your voice in real time — the rate at which your vocal folds vibrate during voiced sound production, expressed in Hertz (Hz). It is specifically configured for the bass and baritone range (80–330 Hz), using a 4096-sample FFT window for higher frequency resolution at low frequencies and an autocorrelation pipeline with a lag search bounded to the deep voice range. The result is displayed in Hz alongside the nearest musical note and a classification reference against bass, bass-baritone, and baritone ranges. For the full technical pipeline, see the How It Works page.
Why is this tool different from a general voice frequency tool? General voice frequency tools are configured for the full vocal range (80–1,100 Hz) and use FFT windows that give adequate resolution for mid and high voices but lose accuracy at the low end. At a 2048-sample FFT window, the frequency resolution is approximately 21.5 Hz per bin — a meaningful error margin for a bass voice at 85 Hz. DeepVoiceTest.com uses a 4096-sample window (resolution ≈10.8 Hz per bin) and constrains the autocorrelation lag search to the 80–330 Hz range. This makes it more accurate and more meaningful for deep-voiced users specifically.
What should I do to get a stable reading? Sustain a low vowel sound — “aah” or “ohh” — at your natural speaking pitch, clearly and steadily. The tool requires at least 100–200 ms of continuous voiced sound for reliable F0 detection. A quiet room, a microphone positioned close to your mouth, and a relaxed, supported tone on a sustained vowel give the best results. Avoid whispering, growling, or phonating at the very bottom of your range where vocal fold closure becomes inconsistent.
Does the tool work on mobile? The tool functions on most modern desktop and Android browsers. iOS Safari has restrictions on AudioContext behaviour that can affect microphone capture for sustained low-frequency analysis. For the most accurate deep voice measurement, desktop use with a wired microphone is recommended.
Understanding Your Result
What is a normal result for a deep voice? As a reference: bass speaking voices typically fall in the 80–150 Hz range; bass-baritone voices in the 90–165 Hz range; baritone voices in the 100–180 Hz range. These are population reference values from acoustic phonetics research. A result below 150 Hz on a sustained comfortable speaking tone indicates a genuinely deep voice by any standard reference. Individual variation within and around these ranges is entirely normal.
My result is 98 Hz. What does that mean? 98 Hz corresponds to approximately G2 in standard equal temperament tuning — squarely within the bass speaking F0 range. It means your vocal folds are completing approximately 98 full vibration cycles per second during phonation at that pitch. For context, the lowest note on a standard piano bass string is A0 at 27.5 Hz; the lowest note typically expected of a trained bass singer in performance is E2 at approximately 82 Hz. A speaking F0 of 98 Hz places your habitual speaking pitch comfortably in bass territory.
My result seems much higher than I expected — around 180–200 Hz. Why? Several factors can cause a higher-than-expected reading. The most common is that the tool detected a harmonic rather than the true fundamental. If your consumer microphone has poor low-frequency response and your true F0 is around 90–100 Hz, the fundamental may be so attenuated in the captured signal that the autocorrelation peaks on the second harmonic (180–200 Hz) instead. Try humming rather than sustaining a vowel — the additional nasal resonance can strengthen the fundamental relative to harmonics. If the result is consistently double what you expect, see the Troubleshooting section below.
My result is around 100 Hz in one test and 200 Hz in another. Why the large difference? A result that is consistently double or half between runs is almost certainly a half-period or double-period detection error — the autocorrelation algorithm found its peak at a lag corresponding to two vocal fold cycles rather than one, or vice versa. This is more common in the deep voice range than in the mid range because the lag values are larger and secondary autocorrelation peaks can compete with the primary. The bounded lag search in this tool reduces this risk, but it is not eliminated entirely. Sustaining a clean, well-supported vowel at a consistent pitch minimises the occurrence of this error.
What does my result tell me about my voice type? It gives you a useful reference point, but not a definitive classification. Voice type — bass, bass-baritone, baritone — is determined by full singing range, passaggio locations (register transition points), vocal timbre, and acoustic quality. Speaking F0 alone does not determine voice type. A baritone with a low habitual speaking pitch and a bass with a relatively high speaking pitch may both produce speaking F0 readings that overlap. Use the result as one data point in understanding your voice, not as a classification verdict.
Accuracy and Limitations
How accurate is the measurement for deep voices? Under typical conditions — a quiet room, a sustained vowel, and a microphone with adequate low-frequency response — the tool achieves F0 detection accuracy of approximately ±5–10 Hz in the 80–200 Hz range. For voices below 100 Hz measured with a built-in laptop microphone, accuracy may be ±10–15 Hz due to consumer microphone low-frequency roll-off. Running multiple tests and noting the consistent range across results gives a more reliable indication of your habitual F0 than any single reading.
Why does my built-in laptop microphone give less accurate results for a deep voice? Consumer microphones — including virtually all built-in laptop microphones — have a frequency response that attenuates signals below approximately 100–150 Hz. This is by design: consumer microphones are optimised for speech intelligibility, which requires clear capture of formant frequencies (300–3,000 Hz) rather than flat response at the fundamental. For a bass voice at 85 Hz, the captured fundamental is significantly weaker than the harmonics, making the autocorrelation peak less distinct. An external USB microphone or headset microphone rated to 80 Hz or below will give substantially more accurate results for very deep voices.
Can this tool measure a voice below 80 Hz? The tool’s autocorrelation lag search is bounded at a maximum lag corresponding to 80 Hz. Voices with a true F0 below 80 Hz — which occurs in some trained bass singers and bass-profundo voices — will fall outside the configured detection range. The tool may return 80 Hz as the floor value or fail to produce a stable reading for sub-80 Hz phonation. This is a documented design choice rather than a technical failure: extending the lag range below 80 Hz significantly increases the risk of noise-driven false positives.
Does background noise affect deep voice detection more than other voices? Yes. The 80–330 Hz frequency range occupied by deep voices overlaps significantly with common ambient noise sources — HVAC systems, traffic, building structure vibration, and some electronic hum. These noise sources compete directly with the voice fundamental in the autocorrelation analysis. Testing in a quiet room is more important for accurate deep voice detection than for mid-range or high voices, where the fundamental sits above most ambient noise.
Deep Voice Science Questions
What makes a voice physically deep? The fundamental frequency of a voice is determined primarily by vocal fold mass, length, and tension. Longer, heavier vocal folds vibrate more slowly under the same subglottal air pressure, producing a lower F0. Adult male vocal folds average 17–25 mm in length; testosterone during puberty drives an increase in fold length and mass that drops the average male speaking F0 from approximately 250–300 Hz in adolescence to 85–180 Hz in adulthood. Bass voices typically have vocal folds at the longer, heavier end of the adult male range.
Can you train your voice to become deeper permanently? The habitual speaking pitch is influenced partly by anatomy and partly by habit, posture, and breath support. Training can lower the habitual speaking F0 by improving posture, relaxing unnecessary laryngeal tension, and developing better breath support — typically by 10–30 Hz in adult males with a consistent practice. However, anatomy sets a floor: vocal fold mass and length cannot be substantially altered through training. The fundamental range a voice can access is largely determined by physiology.
Is a lower fundamental frequency always perceived as a deeper voice? Not always. Perceived voice depth is influenced by F0 but also significantly by formant structure — the resonant frequencies of the vocal tract. A voice with a large pharyngeal cavity and low-set larynx will have lower formant frequencies, contributing to a darker, deeper perceived quality even at the same F0. Two voices at 120 Hz can sound very different in depth depending on vocal tract shape and resonance. F0 is the most measurable component of voice depth; it is not the only one.
About This Website
Who runs DeepVoiceTest.com? DeepVoiceTest.com is built and maintained by Lucas Reed, vocal acoustics specialist and low-frequency voice researcher. Lucas writes all content on the site personally. There are no guest contributors, no sponsored content, and no advertisers. For Lucas’s full background and the reasons this site was built, see the Author page.
Is this site free to use? Yes. The Deep Voice Test and all educational content on DeepVoiceTest.com are completely free. No registration, no account, and no payment of any kind is required.
How do I report an error or ask a technical question? Email contact@deepvoicetest.com with the subject line “Content Correction” for factual errors or “Technical Question” for measurement questions. Lucas reviews all submissions personally. Response time: 48–72 hours.
Written by Lucas Reed, founder of DeepVoiceTest.com. Last updated: June 2026.