Axiom Home Theater Speakers
Get Friendly, Expert Advice
Call 1-888-352-9466 or email
Shipping Included | 30-Day In-Home-Trial

Designed and Manufactured in Canada Since 1980

July 1, 2009

Audio Oddities: Frequency Ranges of Male, Female and Children’s Voices.

Even novices in audio recognize that vibrations of one kind or another are at the root of all sound, and that the frequencies of the vibrations must occur somewhere between 20 Hz and 20,000 Hz in order for humans to hear them through the air as “sound.”

But what is less well understood are the actual frequency ranges of some of the most familiar sounds that each of us hears every day—specifically, men’s and women’s voices, and those of children.

Most of us think that male or female speech or singing voices fall somewhere in the midrange; “midrange vocals” is a recurring phrase commonly seen in speaker reviews, and I’m guilty of using it as well. However, a little research shows that the fundamental frequencies of male and female speech are really much lower in frequency than most of us imagine.

The average man’s speaking voice, for example, typically has a fundamental frequency between 85 Hz and 155 Hz. A woman’s speech range is about 165 Hz to 255 Hz, and a child’s voice typically ranges from 250 Hz to 300 Hz and higher. Of course, each of us has a wider range of sounds that our vocal cords can produce, and if we choose to sing, that range can extend up to four octaves. To help you visualize that, a standard piano keyboard has 88 black and white keys, and covers a frequency range from 27.5 Hz to 4186 Hz, over seven musical octaves (“middle C” is at 261 Hz). One octave represents 12 tones, white and black keys included. We hear the frequency of speech or singing as the pitch, and a doubling of the fundamental frequency represents an octave. If you were singing a note at 100 Hz and then sang it at 200 Hz, you’d hear it as “twice as high.” If you sang a note at 180 Hz and then tried for a really low note, at 90 Hz, the latter would be an octave deeper.

But what’s surprising in all this is that the entire range of men’s and women’s voices remains between about 65 Hz for a male with a very deep bass voice to the highest note of a female coloratura soprano, just above 1,000 Hz, at 1,280 Hz. (A female high-pitched scream can go quite a bit higher, to around 3,000 Hz.)

In loudspeaker terms, looking at a good 2-way speaker like Axiom’s M2 or M3 or M22, that means that the entire range of men’s and women’s singing voices are handled by the woofer before the tweeter even kicks in! Of course there are harmonics of our speech and singing voices that extend higher and give each of us the particular timbre or tonal quality—the harmonic structure– by which we recognize each other’s unique voices, but it still comes as a surprise that the woofer in many speakers, or the woofer and the midrange driver in a 3-way speaker, must handle the full range of human voices, spoken and sung.

This puts an enormous obligation on the speaker designer to make certain that a speaker’s accuracy in the range of the woofer be very linear and smooth, because we’ll certainly use our familiarity with men’s and women’s voices as a criterion of how natural a loudspeaker sounds. If a speaker has a choppy, uneven frequency response in the woofer or woofer/midrange, you can bet it won’t sound natural on spoken or sung male or female voices. Or if a speaker has a big hump or peak between 80 and 150 Hz, it will make male speech and singers sound peculiar—rather chubby or “fat” (a not uncommon design flaw in many speakers). But keeping things linear and smooth throughout that range to 1 kHz, means that the loudspeaker will sound natural on that most familiar of all sounds, men’s and women’s voices (and kids, too).

Alan Lofft was, for 13 years, Editor in Chief of Sound & Vision, Canada’s largest and most respected audio/video magazine. He edited Sound & Vision (Canada) until 1996, when he moved from Toronto to New York to become Senior Editor at Audio magazine.
Lofft has been writing about hi-fi and video professionally for over 20 years, ever since his first syndicated newspaper column, “Sound Advice”, began appearing weekly in The Toronto Star, Canada’s largest-circulation daily newspaper. In the late 1970s, he became a contributing editor, columnist, and equipment reviewer at AudioScene Canada, the leading national consumer electronics magazine at the time. Find out more about Alan in his bio.

Log in


  • August 17, 2013 at 6:20 am

    all must read.

  • Oblomov
    April 23, 2015 at 6:41 am

    “The average man’s speaking voice, for example, typically has a fundamental frequency between 85 Hz and 155 Hz. A woman’s speech range is about 165 Hz to 255 Hz, and a child’s voice typically ranges from 250 Hz to 300 Hz and higher.”
    As is specified later on in the article, there is much overlap.
    As these intervals might make it look (or sound?) like there is not overlap in the speaking range of men and women, I assume it talks of a reference average frequence taken out of many, not just a single man.
    85 – 155 is also in fact less than an octave range (85 would need to double up to 170) so i assume your reference is the average speech pattern of many voice types, or “simply” an estimated “fundamental frequence” of their chords at a given state of tension/relaxation approximately used as a reference, as we know pitch is inversely proportional to mass and and directly to tension.
    Then 85 hz, roughly F2, would be a deep bass relatively relaxed voice, 155 hz, roughly around D#3 would be that of a relaxed high tenor. While 165, between E and F3 would be a low alto relaxed speech, while 255 approaches the middle C, which is around the lowest notes of soprano tessitura, but not the real relaxed range of any but the highest soprano voices. So this seems to confirm, that men and woman ranges are actually closer than what classical tessitura seem to indicate, and between tenors and soprano there is far less than an octave of difference, women just are allowed to reach their upper register, but (double standard) that is considered falsetto for men, but baritone countertenors use it, it’s strong and a legitimate part of their voice, and althought some actually use a reinforced falsetto (especially for chamber baroque music, which has a much lower orchestra), many just use their full upper register (if your idea of “full” man voice is a tenor, that’s ok, but tenors try to take their middle register up to C-D5, or they try to twang and shove as much of their low chest consonance into their early upper register), some still sense a difference between their upper registers and that of woman and call it falsetto, but I’d call it different placement, the same difference between a F#5 in the head voice of an Alto and that of a soprano, you sense one is a lower voice, because of subtly different overtones.

  • James Walter Taylor
    May 4, 2015 at 12:31 pm

    The difference between specifying a fundamental frequency vs. median frequency or peak energy and a bandwidth for clear reproduction quite different – It’s good to keep all of these factors front and center. Oddities like long transient tails on sharp cut-off filters which are not necessarily dependent on the cut-off frequency are also important to clear reproduction. Alveolar sibilants have a huge bandwidth to remain uncolored.

    Telephone bandwidth 300-3400Hz – at a guess a minimum analog 3dB band of 80-16KHz, with a relatively constant group delay, gives you good, undistorted, intelligible speech.

  • John Picton
    August 14, 2016 at 3:23 am

    Thank you, this information was particularly useful for notes I am preparing for the sound team at our Church

    • August 15, 2016 at 5:26 am

      Glad it helped, John!

RSS feed for comments on this post.

Leave a comment

клиентские онлайн игрыбраузерные онлайн игры