Lucas Reed
Vocal Acoustics Specialist & Low Frequency Voice Researcher | Founder, DeepVoiceTest.com
There is a specific moment most people with deep voices encounter at some point — a recording plays back, someone says “your voice is unusually deep,” and the question surfaces: how deep, exactly, and what does that mean acoustically? The answer is measurable. The tools to measure it clearly and explain what the number means have been, until now, surprisingly hard to find.
My name is Lucas Reed. I have spent years researching the acoustic science of low-frequency voice production — specifically, the physiological and acoustic factors that produce a deep voice, how fundamental frequency at the low end of the human range is detected and measured accurately, and why consumer microphones and standard browser audio tools often underperform precisely in the frequency range where deep voices live.
A bass voice has a fundamental frequency (F0) as low as 80 Hz. At 44,100 Hz sample rate, detecting a 80 Hz fundamental accurately requires an FFT window large enough to resolve that frequency — at minimum 4096 samples, giving a frequency resolution of approximately 10.8 Hz per bin. Standard pitch detection tools configured for the full vocal range (80–1,100 Hz) often sacrifice low-frequency resolution in favour of broader coverage. DeepVoiceTest.com is built specifically for the low end — where the resolution, the reference ranges, and the explanations are calibrated for voices that actually live below 200 Hz. To understand exactly how the detection pipeline works for low-frequency voices, see the How It Works page.
What I Research and Write About
Low Fundamental Frequency Detection and Measurement Detecting F0 accurately below 120 Hz presents specific technical challenges that mid-range pitch detectors are not optimised to handle. Consumer microphones roll off at low frequencies. Autocorrelation algorithms can produce half-period errors — returning 100 Hz when the true F0 is 50 Hz — more frequently in the low vocal range than in the mid range. I research these failure modes specifically and document how DeepVoiceTest.com addresses them, including appropriate FFT window sizing, autocorrelation lag range configuration, and the role of microphone frequency response in low-F0 accuracy.
Vocal Fold Physiology and the Acoustic Basis of Deep Voice Deep voices are produced by longer, heavier vocal folds that vibrate more slowly under subglottal air pressure. Adult male vocal folds average 17–25 mm in length; bass singers typically have folds at the longer end of this range. Testosterone during puberty drives the dramatic F0 drop — from approximately 250–300 Hz in adolescence to 85–180 Hz in adult males — by increasing fold length and mass. I write about the physiology behind these changes because understanding why a voice is deep is as important as measuring how deep it is.
Voice Type Classification for Deep Voices The classical voice classification system — bass, bass-baritone, baritone — is based on range, timbre, and register transition points (passaggi), not on a single F0 measurement. However, speaking F0 is a measurable and informative starting point. I document the relationship between measured speaking F0 and traditional bass/baritone classification, where those categories begin and end in Hz terms, and why a single measurement cannot definitively classify a voice type. The FAQ covers the most common questions about result interpretation for deep voices.
Deep Voice Acoustics in Media, Broadcasting, and Performance Bass voices are disproportionately represented in certain professional contexts — voice acting, radio broadcasting, film narration, operatic performance. The acoustic characteristics that make a voice sound authoritative, resonant, or commanding are partly a function of F0 and partly a function of formant structure, vocal tract resonance, and subglottal coupling. I research and write about these dimensions for the audience of working voice professionals and aspiring performers who want a scientific basis for understanding their instrument.
Why I Built This
When I was looking for a browser-based tool that could accurately measure fundamental frequency in the bass vocal range — below 120 Hz — and explain what that measurement meant in terms of voice type classification, I found tools designed for the full vocal range that lost accuracy precisely at the low end. The typical consumer microphone frequency response chart tells the story: most microphones are flat from 200 Hz upward, but roll off below 150 Hz. A tool that does not account for this will systematically underperform for the users who need low-frequency accuracy most.
What I built is a tool calibrated specifically for deep voice measurement — with an FFT configuration optimised for low-frequency resolution, documented accuracy figures for the bass and baritone range, and reference ranges drawn from acoustic phonetics research rather than assumed from general pitch detection norms.
Accuracy Standards
Every frequency range, algorithm description, and voice science claim on DeepVoiceTest.com is cross-referenced against published acoustic phonetics literature, vocal physiology research, and the W3C Web Audio API specification. Low-frequency detection accuracy figures are documented with specific reference to the measurement conditions under which they apply — including microphone type, FFT configuration, and the effect of OS audio processing on low-frequency signal capture. Limitations are stated explicitly and prominently, not buried. For how your audio is handled during testing, see the Data Security page.
Tools on This Site
- Deep Voice Test — deepvoicetest.com
Get in Touch
For content corrections, technical questions, or data requests, visit the Contact page or email directly: contact@deepvoicetest.com
Response times: technical and content questions within 48–72 hours. Privacy and data requests within 7 business days. GDPR requests within 30 days. CCPA requests within 45 days.
Lucas Reed is the founder and sole author of DeepVoiceTest.com. For a full account of how content is researched and written, see the Editorial Guidelines. Last updated: June 2026.