The Science of the Singing Voice

1
The Science of the Singing Voice

• Overview of the course (HC16)
• Winter 2006
• Pat Keating, Linguistics, UCLA

2
Books

• Johann Sundberg, The Science of the Singing
  Voice. Northern Illinois University Press (1989)
• Peter Ladefoged, Elements of Acoustic Phonetics.
  Second edition. University of Chicago Pres
  (1996)
• Richard Miller, The Structure of Singing System
  and Art in Vocal Technique. Wadsworth Publishing
  (2001)
• Richard Miller, National schools of singing
  English, French, German, and Italian techniques
  of singing revisited. Scarecrow Press (2002)
• Garyth Nair, Voice Tradition and Technology A
  State-of-the-Art Studio. With CD. Singular
  (1999)

3
Intro Sundbergs demo

• Go to The ugly voice poster

4
I. Digital audio files

• Audacity tutorial on digital audio
• Ripping CD tracks to .wav (Real, CDex)
• Saving .mp3 as .wav (Audacity,Cdex)
• Splitting and saving tracks from stereo
  (Audacity)
• (go to CDEx, Audacity)

5
Our week 1 labA Sound Library of clips

• Making 1-channel .wav files lt 30 sec
• Go to Sound Library on web (logged in as a
  student)

6
II. Pitch

• Frequencies of musical notes each doubling of
  frequency is an octave
• Semi-tone almost 6
• in tune how close is close enough (20 cents?)
• in tune steadiness
• Vibrato vs straight tone

7
Vibratos

• Dimensions of vibrato
• Rate, range, amplitude vibrato
• Supposed good vibrato
• 5.5 to 7 Hz, .5 to 2 semitones
• What good a vibrato does, doesnt do for the
  singer
• Examples of vibrato from classical, pop

8
Pitchtracking

• Hardest part keeping track of F0 range
• Tuning forks and thin voices dont use cepstral
  method (sample file of tuning fork)
• Speech Analyzer 500 Hz limit
• Problems tracking trills changing step size and
  window length

9
Our week 2 lab F0

• F0 matching
• F0 steadiness
• Measuring vibratos

10
III. Spectrum

• A bit on laryngeal anatomy and mechanism of
  vibration
• The voice source F0 and overtones
• Line spectrum of source, FFT of output
• DVD Human Speech speed of closing determines
  strength of higher harmonics

11
Partials, overtones?

• Partials harmonics
• Overtones partials above F0

12
Our week 3 lab FFT

• FFT, LTAS in Pitchworks
• FFT in Audacity View-Plot spectrum (nice for
  comparing window lengths)
• pros and cons of Audacity/Pitchworks
• Comparing spectra of different voice qualities
• Looking at strength of H1, number of harmonics,
  amount of high-freq energy

13
IV. Recording the source

• Sundberg All about the flow glottogram (Ug),
  from inverse filtering of Uo signal

14
2 key aspects of the flow glottogram

• the maximum amplitude of the flow is directly
  proportional to the amplitude (in the source, not
  in the output) of the fundamental component
•     and this affects the perceived strength of
  the voice, though not necessarily its overall
  loudness, which instead depends on the strongest
  partial
• the maximum closing rate is proportional to the
  amplitudes of the overtones

15
Pressed, breathy, flow phonation

• pressed phonation high lung pressure combined
  with adducted glottis the adducted glottis
  requires more pressure to get vibration, but
  still, little air flows through the narrow and
  brief opening low airflow low amplitude of the
  flow glottogram, so a weak F0 component.
• breathy phonation glottis is somewhat abducted
  so there is never complete closure this means
  not only that some air flows through
  continuously, but also that the maximum flow is
  quite high high airflow a strong F0 component,
  but also a noise component in the voice (usually
  seen instead of higher-frequency partials)
• flow phonation not so high lung pressure, and
  the most abducted glottis that will still give
  complete closure. This means the greatest
  possible amplitude of the F0 component without
  the noise. The amplitude of the F0 in flow
  phonation can be 15 dB or more greater than in
  pressed phonation.

16
A different view of the source EGG

• Ch. 13 in Nair (1999) The Use of the
  Electroglottograph in the Voice Studio by D.
  Miller and H. K. Schutte
• one of the primary aims of training the
  classical singing voice will be to establish the
  habit of complete and abrupt closure, at least in
  mezzo forte and forte
• Their sample sound file on next slide

17
Falsetto vs chest voice on i little contact in
falsetto
18
Our week 4 lab EGG

• EGG recording of each individual student,
  channels then split into separate .wav files
• Listen to EGG signal
• Spectrum of EGG signal
• Compare shape of pulses to examples

19
V. Resonances

• A bit of source-filter theory and vowel formants,
  including from DVD
• Singers formant extra energy around 3000 Hz
  (Sundberg says 2300-3000 Hz for basses, 3000-3800
  for tenors), which allows a solo voice to stand
  out against an orchestra, or other singers
• (Sopranos dont much need a singers formant
  against an orchestra, because any note above
  about B4 will stand out by itself. Similarly for
  amplified singers.)

20
Miller singers formant
21
Singers formant

• Not really an additional formant, but a
  clustering of F3, F4, F5 when they are close
  together in frequency their strengths are
  mutually enhanced, giving one broad strong
  spectral peak.
• Male singers enlarge the ventricle (just above
  the larynx), lower the larynx.
• It is not known how altos produce their singers
  formant.
• But I made a big difference in my voice by
  following male instructions

22
Speakers formant

• More like at 3500 Hz than 3000
• Property of speaking voices judged good
• Seen in some singing voices, especially in styles
  that are more like speaking

23
Our week 5 lab resonances

• Looking at own voice and at recordings for
  singers formant trying to increase singers
  formant
• Comparing vowels for the effect of different
  formant frequencies on strengths of different
  harmonics, esp., on strength of H1

24
VI. F1 tuning

• a strong voice matches H1 to F1, while a weak
  voice has no formant near H1
• Good illustration of this on DVD the good voice
  and the bad voice samples
• Sundberg says that tuning H1 to F1 can add up to
  30dB to the sound level.
• F1 is raised by opening the mouth more, or
  shortening the vocal tract (e.g. smiling)

25
When F0 is above F1

• F0 gt F1 for many soprano notes
• F1 cannot match F0, and also vowel qualities are
  indistinct
• trained singers tend to adjust the vowel quality
  so that the F1 moves up, in the direction of F0.
  
• (Is this what I do? No, I generally have F1
  tuned to H2 - see later slide.)

26
Sundberg F1 tuning when F0gtF1
27
The soprano challenge

• Recently a study of this effect, explicitly
  testing what Sundberg had said, got a lot of
  publicity
• http//www.phys.unsw.edu.au/jw/soprane.html
  (this page saved to computer but without sound)
• They found that a trained soprano singing above
  about 440 Hz tuned every vowels F1 to the F0.

28
Miller on passagio tuning

• The crucial point in a correct execution of
  passaggio is to avoid pushing the chest register
  beyond its natural limits by means of a forced,
  shouty production. This tendency, natural in
  most untrained voices, appears as a compulsion to
  tune the first formant to the second harmonic of
  the voice source in the attempt to extend the
  chest register upward. Singing teachers
  recognize this phenomenon in the undesirable
  raising of the larynx when approaching the upper
  range. this increases the frequency of F1
  The larynx raising is thus a maneuver of last
  resort to get the first formant higher in order
  to add a semitone or two to the chest register.
  (...) The typical high note which is forced in
  this way has a second harmonic with a level 15 to
  25 dB higher than the first harmonic. (...) One
  reason for the strong tendency to tune F1 to H2
  on high notes in the chest register is that the
  resulting resonance is quite powerful. (...) A
  compensatory adjustment for this loss that most
  accomplished (opera) singers employ is the tuning
  of F2 to a higher harmonic.

29
Our week 6 lab F1 tuning

• Mapping out a singers pitch range and F1 range
  which vowels have F1 in the overall F0 range, and
  which vowels would be best to sing on a given
  pitch to get F1 near that F0 (e.g. if I can sing
  from A220 to A880)
• Trying to tune F1 to F0
• Passaggio out of chest voice

30
VII. Breathing in singing

• Recall
• Muscles that can participate in inhalation
  (expansion) external intercostals, diaphragm
• Muscles that can participate in exhalation
  (contraction) internal intercostals, abs

31
Breathing in singing

• Normal breathing about .5 liters 12
  times/minute, with active inspiration and passive
  expiration.
• Singing much longer breaths, and more total air
  in a breath. More of the air in the lungs is
  exhaled by professional singers.

32
Subglottal pressure

• Pressures generated in speech are much lower than
  those in singing, but even singing pressures are
  less than those used by reed and brass instrument
  players, which in turn are less than those used
  in lifting heavy weights (with the glottis
  completely closed).
• In singing, pressure is higher for louder
  phonation and for higher pitches A doubling of
  subglottal pressure gives about a doubling in
  loudness, and subglottal pressure also about
  doubles when F0 doubles.

33
Oral flow

• The flow of air out of the mouth depends on the
  pressure in the lungs, and the size of the
  opening through the larynx and vocal tract.
• In pressed phonation, subglottal pressure is
  high, but the glottis is nearly closed so airflow
  is low. In breathy phonation, the more-open
  glottis lets air leak out throughout the glottal
  cycle.
• Airflow tends to increase for louder or
  higher-pitched notes.
• Vibrato uses more air than straight tones.

34
Trained vs. untrained

• Classically trained singers have lower subglottal
  pressures than do untrained singers, and these
  pressures are lower in speech as well as singing.
• Trained singers have lower airflow rates in
  singing than do untrained singers, but the same
  airflow rates in speech.
• Trained singers thus have more efficient
  phonation they use less air to get strong vocal
  fold vibrations.

35
Loudness control with

• a. phonation the right amount of vocal fold
  adduction (Sundbergs flow phonation)
• b. the vocal tract formant tuning, singers
  formant
• c. lung pressure higher pressure and higher
  airflow through the glottis. The power of the
  glottal source increases by 6 dB for every
  doubling of the lung pressure

36
Sundberg Ps vs. pitch
sound level (S), subglottal pressure (P) and oral
airflow (A) from a professional singers
ascending scale, showing how pressure increases a
lot as pitch increases, even when airflow is
fairly constant and sound level increases only
somewhat
37
Sundberg Ps vs. pitch
the clear relation of loudness, pressure and
pitch in these quicker triads
38
Our week 7 lab aero

• Pressure and flow for each student
• Did they show the relation of Ps, Uo, and F0
  (with relatively fixed loudness) as in the
  Sundberg example figure?

39
VIII. Consonants

• 2 chapters each in Miller, Nair
• Nair says that trained singers have greater
  amplitude, more energy in more harmonics, and
  stronger resonances in their consonants, and that
  this helps smooth connections between consonants
  and vowels.
• Nairs resonance checklist for consonants jaw
  down, proper tongue configuration, raised
  velum, pharynx open and relaxed, larynx relaxed
  and low in the throat. No extreme lip spreading
  or rounding, because both tend to raise the jaw.

40
Consonants

• F0 perturbation effects from consonants
• Miller is mostly concerned with agility in
  consonant articulation so that consonants can be
  made quickly and take up little time relative to
  vowels, and he includes agility exercises
• Nair mentions lengthening of glides for dramatic
  effect, advises minimal noise for word-final
  consonants

41
Our week 8 lab consonants

• Comparing sung consonants for relative
  vowel-likeness, speed
• Pitch perturbations
• Dramatic effect of lengthening, releasing, etc.
  of consonants

42
IX. Comparing singing styles

• Millers book on western classical traditions
• Several comparisons of genres in the literature
  indicate that on various measures, pop,
  Country-Western and Broadway singing are more
  like speaking than is classical (operatic)
  singing
• Vocal tract in classical singing has wider lower
  pharynx, wider oral cavity

43
Our week 9 lab final projects

• Most students are comparing song samples, either
  commerical or recorded by them
• Projects are being constructed as webpages
• Easy to combine presentation of text, graphics,
  sound materials
• When done, can be viewed by whole class
• Class on using Dreamweaver by ITC

44
X. Titze on warm-ups

• Titze explains warm-up exercises in terms of
  bringing all systems up gradually
• Acoustic loading for respiratory warm-up
• increase the acoustic loading on the vocal folds
  with humming, trills, singing into a straw - lets
  the vocal folds vibrate with more abduction, and
  with overall lower Ps for an easy start
• increase F0 so that Ps must increase