Overtone singing

« previous post | next post »

Anna-Maria Hefele:

In the first "scale" that she sings, the fundamental frequency is about 273 Hz (which is a bit flatter than C#4), and as the spectrogram below shows, the "scale" then picks out the 4th, 5th, 6th, 7th, 8th, 9th, and 10th harmonics (confusingly, these are the 3rd, 4th, 5th, 6th, 7th, 8th, and 9th overtones, since the fundamental frequency is the 1st harmonic but the 0th overtone):

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

So the frequencies of the "scale" are



1092   1365   1638   1911   2184   2457   2730

Clearly she is creating a relatively narrowly tuned vocal-tract resonance that is able to enhance a single harmonic significantly above its neighbors (though looking at spectral slices suggests that there are also apparently some anti-resonances involves)




  1. bks said,

    October 10, 2014 @ 9:38 am

    I initially read the headline as Overtone Signing which would have been equally interesting, and watching her hand motions, equally apposite! –bks

  2. Mark F. said,

    October 10, 2014 @ 10:04 am

    This is a dumb question, but it just occurred to me that I don't know the answer — in resonant amplification, where does the energy come from? Other wavelengths?

  3. Brett said,

    October 10, 2014 @ 11:05 am

    @Mark F.: One way of thinking of resonant enhancement is that it involves tuning the phase of the system so that the periodic forcing (i.e. the compression of air in the voicebox by the vocal chords) is most efficient in transferring energy to the output oscillations (i.e. the density oscillations in the air). When the forcing function and the forced oscillations are out of phase, little work is actually performed by the force; when they are in phase, almost all the possible work is extracted from the forcing.

  4. Howard Oakley said,

    October 10, 2014 @ 11:22 am

    Wonderful, thank you. There are some parts of the world where this is not so unusual – I seem to recall that the Caucasus is one (in addition to their polyphonic singing), but might be incorrect by a few thousand miles!
    I think that most of the tuning to achieve the overtones is performed in the mouth, rather than closer to the vocal chords, but stand to be corrected there too, I suspect!

  5. George Grady said,

    October 10, 2014 @ 12:45 pm

    As impressive as that is, I wouldn't generally want to call it singing, as there aren't any words. It seems more akin to humming or whistling (or humming and whistling?). Does anyone know if it's possible to do that sort of thing while singing actual words?

    [(myl) Because the "overtone singing" technique requires shaping the vocal-tract resonances to capture some particular overtone in the 800-3000 Hz region (or so), and because the articulation of vowels and consonants requires essentially the same thing (though with more broadly-tuned resonance areas), I suspect it would be quite hard to do both at once in a general way. See e.g. Gerrit Bloothooft et al., "Acoustics and perception of overtone singing", JASA 1992:

    Overtone singing, a technique of Asian origin, is a special type of voice production resulting in a very pronounced, high and separate tone that can be heard over a more or less constant drone. An acoustic analysis is presented of the phenomenon and the results are described in terms of the classical theory of speech production. The overtone sound may be interpreted as the result of an interaction of closely spaced formants. For the lower overtones, these may be the first and second formant, separated from the lower harmonics by a nasal pole‐zero pair, as the result of a nasalized articulation shifting from /c/ to /a/, or, as an alternative, the second formant alone, separated from the first formant by the nasal pole‐zero pair, again as the result of a nasalized articulation around /c/. For overtones with a frequency higher than 800 Hz, the overtone sound can be explained as a combination of the second and third formant as the result of a careful, retroflex, and rounded articulation from /c/, via schwa /E/ to /y/ and /i/ for the highest overtones. The results indicate a firm and relatively long closure of the glottis during overtone phonation. The corresponding short open duration of the glottis introduces a glottal formant that may enhance the amplitude of the intended overtone. Perception experiments showed that listeners categorized the overtone sounds differently from normally sung vowels, which possibly has its basis in an independent perception of the small bandwidth of the resonance underlying the overtone. Their verbal judgments were in agreement with the presented phonetic‐acoustic explanation.


  6. Carrington Dixon said,

    October 10, 2014 @ 2:08 pm

    Singing without words is call vocalise see http://en.wikipedia.org/wiki/Vocal_warm_up#Vocalise. It's not exactly common in classical and jazz, but it's not all that rare either.

  7. Captain Bringdown said,

    October 10, 2014 @ 2:14 pm

    Howard Oakley,

    You're likely thinking of Mongolian throat singing: http://en.wikipedia.org/wiki/Tuvan_throat_singing

  8. maidhc said,

    October 10, 2014 @ 3:08 pm

    Glen Campbell used to do an overtone-singing version of "Bonaparte's Retreat" when he was a guest on various TV talk shows. I've been able to find a clip of it though.

  9. D-AW said,

    October 10, 2014 @ 4:10 pm

    This singer is wonderful. Among other things of interest is the comparison between her Western, quasi-classical technique and traditional overtone singing (Tuvan, which enjoyed some popularity in the early 2000s, but also (other) Mongolian styles, Buddhist tantric chanting, even Inuit game songs). Tavener's opera Mary of Egypt has a part for overtone singer, but it doesn't take advantage of the kind of range demonstrated here. It's used there mostly for atmospheric effect.

  10. Roger Lustig said,

    October 10, 2014 @ 4:42 pm

    Amazing indeed.

    Just to be a grouch, that's not a "folk song" as she says–it's by Mozart, K.596, Komm', lieber Mai und mache. I guess it's become a folk song over time…but that flatted 7th at the end is Anna-Maria's own.

  11. Rubrick said,

    October 10, 2014 @ 4:44 pm

    Amazing how all it takes for a video to go viral is for it to feature someone who is immensely talented, highly trained, and stunningly beautiful. (Thanks for posting.)

  12. Stan Carey said,

    October 11, 2014 @ 4:32 am

    Look up Huun-Huur-Tu or Yat Kha on YouTube for excellent Tuvan throat-singing. Huun-Huur-Tu are folky and traditional sounding; Yat Kha are more in a contemporary vein.

  13. tpr said,

    October 11, 2014 @ 5:18 pm

    There are some nice examples of Tuvan throat singing in this documentary about Richard Feynman too (particularly at 28m48s and 33m28s).

  14. AntC said,

    October 11, 2014 @ 6:56 pm

    Thank you Mark for the find, certainly a very accomplished technique.

    As to its musicality, I'm unconvinced — an acquired taste, like Vogon poetry(?) A propos of which …

    @Carringon Dixon, yes Vocalise is singing without words; noting scat singing http://en.wikipedia.org/wiki/Scat_singing it's probably more accurate to call Vocalise singing without syllables; and Mendelssohn's Lieder Ohne Worte shows there can be singing without voice; but I think this overtone stuff shows that not all singing without words/syllable is Vocalise.

  15. Jason Stokes said,

    October 11, 2014 @ 9:02 pm


    A pretty girl doesn't even have to talented to become Internet famous. Pardon me for being cynical, but there are dozens of better throat singers on Youtube and the one that gets posted is the featuring a pretty girl. So you might say, amazing how all that it takes for a video to go viral is it to be novel and feature a pretty girl.

    [(myl) Please avoid contentless snark of this kind in our comments section. You claim that there are "dozens of better throat singers" without telling us who one of them is, or by what measures they are better. And I have not seen a clip that demonstrates the technique of picking out different harmonics as clearly and straightforwardly as this one does, which is why I analyzed and posted it.

    Since someone responded to you before I saw your comment, I won't delete it — but otherwise I would have done so.]

  16. Per said,

    October 12, 2014 @ 2:04 pm

    ugh @Jason Stokes

    Did you watch anything else she's done? Pardon me for being cynical, but it's amazing how all that it takes for men on the internet to dismiss a pretty girl's talent is for her to make an instructional video that isn't designed to showcase her virtuosity.

  17. ron said,

    October 14, 2014 @ 9:55 pm

    Jim Seals (of Seals & Crofts) was known for performing Bonaparte's Retreat this way. You can hear him demonstrate his version of the traditional tune in a 1971 radio concert preserved at Internet Archive (https://archive.org/details/pra-BC0709.02). He introduces it around 19:10, & the actual overtone performance starts about 20:25.

RSS feed for comments on this post