Logopedics Phoniatrics Vocology, 2012; Early Online: 1–6

1 57
2 58
3 lpv forum 59
4 60
5 61
6 62
7 Acoustic passaggio pedagogy for the male voice 63
8 64
9 65
10 66
11 Kenneth Wood BOZEMAN 67
12 68
13 Lawrence University, Conservatory of Music, 711 E. Boldt Way, Appleton,Wisconsin 54911, USA 69
14 70
15 71
16 72
17 Abstract 73
18 Awareness of interactions between the lower harmonics of the voice source and the first formant of the vocal tract, and of the 74
passive vowel modifications that accompany them, can assist in working out a smooth transition through the passaggio of the
19 male voice. A stable vocal tract length establishes the general location of all formants, including the higher formants that form
75
20 the singer’s formant cluster. Untrained males instinctively shorten the tube to preserve the strong F1/H2 acoustic coupling 76
21 of voce aperta, resulting in ‘yell’ timbre. If tube length and shape are kept stable during pitch ascent, the yell can be avoided 77
22 by allowing the second harmonic to rise above the first formant, creating the balanced timbre of voce chiusa. 78
23 79
Key words: Close timbre, first formant, harmonics, open timbre, passaggio, passive vowel modification, singer’s formant cluster,
24 voce aperta, voce chiusa, voice source 80
25 81
26 82
27 Introduction and overview trait. Yelling results in an increasingly pressed 83
28 mode of phonation, a heavier (thyro-arytenoid 84
29 A relatively stable tube (vocal tract) length is nec- 85
(TA)-dominant) laryngeal registration adjustment,
30 essary for consistent depth and balance of timbre 86
and inconsistency of timbre and perceived Fach. If
31 in Western classical singing. Both tube length and 87
vocal tract length and shape are kept stable, F1
32 cross-sectional area are responsible for formant 88
locations will remain stable and the yell will be
33 locations. Effective tube length determines the 89
avoided by allowing H2 to pass through and above
34 general location of the entire formant set of a given 90
F1. This transition results in the timbral shift var-
35 voice. Formants one and two (F1 and F2) are most 91
iously referred to as ‘covering’ or ‘turning over’,
36 affected by changes of shape other than length 92
and it facilitates a lighter laryngeal registration
37 (changes in cross-sectional area) and thereby effec- 93
adjustment and a less pressed mode of phonation
38 tively define and differentiate vowels. Higher for- 94
(4–6). Since the location of the first formant varies
39 mants are less responsive to changes of shape other 95
by vowel, knowledge of the variety of F1 locations
40 than length. Tube length is therefore a significant 96
and their predictable F1/H2 crossings an octave
41 factor in setting and maintaining the frequency of 97
lower becomes both an accurate means of assess-
42 higher formants. These higher formants are drawn 98
ing tube length stability and a reliable basis for
43 together to form the singer’s formant cluster (SFC) 99
developing effective strategies for training tube
44 under certain conditions (low larynx, open throat, 100
stability for successful negotiation of the passaggio.
45 and narrow epilaryngeal tube) as described by 101
46 Sundberg (1) and refined by Titze and Story (2). 102
47 The relative stability of the SFC is a key element 103
Acoustic principles
48 in consistency of timbre and of perceived vocal 104
49 Fach. The pedagogic problem: Untrained males Timbral transitions once thought to be the result of 105
50 instinctively shorten the vocal tract upon ascend- laryngeal muscular adjustments are caused instead 106
51 ing, raising F1 in order to preserve the strong F1/ by changing interactions between harmonics of the 107
52 H2 acoustic coupling of the ‘yell’ (3), a behavior voice source and the first two formants of the vocal 108
53 that appears to be a universal, hardwired survival tract. This is not to say that there is no source 109
54 110
55 Correspondence: Kenneth Wood Bozeman, Lawrence University, Conservatory of Music, 711 E. Boldt Way, Appleton, Wisconsin 54911, USA. E-mail:
111
56 [email protected] 112
(Received 8 January 2012; accepted 4 March 2012)
ISSN 1401-5439 print/ISSN 1651-2022 online © 2012 Informa UK, Ltd.
DOI: 10.3109/14015439.2012.679967

SLOG_A_679967.indd 1 4/16/2012 1:05:45 PM

 2 K. W. Bozeman
1 contribution to timbre, but the particular shifts of the •• Nota bene: Any vowel can be executed in open 57
2 male passaggio are primarily caused by harmonic/ or close timbre, depending upon range, or pro- 58
3 formant interactions. There is an audible effect when nounced more openly or more closely by raising 59
4 a harmonic passes through a formant, especially or lowering its F1. 60
5 when a lower harmonic passes through formants one •• Chiaroscuro timbre: a balanced tone with both 61
6 or two. The most notable of these effects occurs when ring and depth, usually accomplished by some 62
7 H2 passes through F1, traditionally described as balance in power between F1 and a higher 63
8 moving from open to close timbre, turning over, or formant, either the singer’s formant cluster or 64
9 covering. If tube length and shape are kept stable F2, depending upon the situation or range. 65
10 while harmonics rise, this H2/F1 crossing will occur •• Turning over: the sound and sensation of H2 66
11 at predictable locations within a given voice. While slipping above F1; also called covering, closing, 67
12 each voice is unique, formant locations are rather tipping, etc. 68
13 predictable per vowel within a given vocal Fach, usu- •• Yell (call, shout): an open-timbred F1/H2 cou- 69
14 ally to within a half step. Indeed, formant frequency pling carried above the normal F1 location of 70
15 locations (assuming a stable tube length) are signifi- the vowel being sung (Figure 2). 71
16 cant in determining vocal Fach. 72
17 Intermezzo: yell characteristics 73
18 74
19 Cardinal vowels’ approximate first formants The yell is accomplished by raising F1 through tube 75
20 and locations of turning (Figure 1) shortening (larynx raising), pharynx narrowing, and 76
21 mouth widening. It tends toward pressed phonation, is 77
Definitions
22 typically done on open vowels like /ż/ or /Ɔ/, and is chest 78
23 The principles at play in acoustic passaggio pedagogy register (TA)-dominant. This shortened, divergent res- 79
24 are inherent in the following definitions: onator shape is typical of most popular, folk, and world 80
25 music, which can be described as more or less skillful 81
•• Open vowel: a vowel with a high F1, such as /ż/, yelling. It is clear and powerful but typically involves
26 82
/Ɔ/, /ſ/, with an inherently divergent resonator higher levels of breath pressure and glottal resistance.
27 83
shape. (A divergent resonator shape is one in
28 84
which the resonator is increasingly larger or •• Whoop (hoot): a strong F1/H1 coupling (3)
29 85
more open at the lip end.) (Figure 3).
30 86
•• Close vowel: a vowel with a low F1, such as /i/
31 87
or /u/, with an inherently convergent resonator
32 Intermezzo: whoop characteristics 88
shape. (A convergent resonator shape is one in
33 89
which the resonator is increasingly narrower or The whoop or hoot is accomplished by tuning F1 to the
34 90
closer at the lip end.) H1 (F0) being sung. It is falsetto (cricothyroid)-domi-
35 91
•• Open timbre (voce aperta): sounds in which two nant, tends toward flow phonation, facilitates laryngo-
36 92
or more harmonics are at or below the F1 of the pharyngeal space and a convergent resonator shape
37 93
vowel being sung. (both of which lower F1), and is primarily done on close
38 94
•• Close timbre (voce chiusa, covered, turned over): vowels, especially /u/. Since women often sing pitches
39 95
sounds in which H2 has slipped above F1 of the above the F1 locations of their vowels, they are often
40 96
vowel being sung. operating in whoop mode in Western classical singing.
41 97
42 98
43 99
44 100
45 101
46 102
47 103
48 104
49 105
50 106
51 107
52 108
53 109
54 Figure 1. This is a ‘musician-friendly’ charting of formants. In this chart, the approximate first formant locations are notated on the treble 110
clef with IPA symbols within boxes, while the pitches on the bass clef indicate the pitches above which each vowel would ‘close’ or ‘turn
55 over’. The historic location of the zona di passaggio as reported by Richard Miller (7), lies between the turning frequencies of the moderately
111
56 close /e, o/ and the most open /ż/ vowels. 112

SLOG_A_679967.indd 2 4/16/2012 1:05:56 PM

 Vocal pedagogy  3
1 57
2 58
3 59
4 60
5 61
6 62
7 63
8 64
9 65
10 66
11 67
12 68
13 69
14 70
15 Figure 2. F1/H2 coupling of open timbre. 71
16 72
17 •• Vowel modification: an adjustment of a vowel (a 2. When any source harmonic descends through 73
18 retuning of F1 and F2 locations) accomplished F1, there is a noticeable timbral opening. 74
19 by a reshaping of the vocal tract. 3. When any source harmonic ascends through F1, 75
20 •• Passive vowel modification: a vowel shift that there is a noticeable timbral closure; the fewer 76
21 occurs not from a shape change, but from the harmonics remaining the below F1, the greater 77
22 changing interactions between a stable tube the closure. 78
23 shape (and formants) and rising pitch (rising 4. When H2 approaches the peak of F1, it forms 79
24 source harmonics). a strong, open timbre acoustic coupling. 80
25 If this F1/H2 coupling is taken higher 81
26 through tube shortening, this becomes a shout 82
27 F1/harmonic interaction summary 83
or yell.
28 Awareness of degrees of openness and closeness is 5. Conversely, when H2 is allowed to pass through 84
29 crucial to understanding male passaggio. Vowel and and above F1, the primary timbral transition of 85
30 timbral openness or closeness is related to the height the voice (i.e. turning over) occurs. 86
31 of the first formant frequency and/or to the relation- 6. An F1/H1 coupling (an octave higher than F1/ 87
32 ship between lower source harmonics and the first H2) becomes whoop or hoot timbre and is 88
33 formant. It is therefore pedagogically necessary to ‘heady’ and full. 89
34 understand the relationships between the first 7. As one ascends the octave between turning and 90
35 formant and the harmonics of the voice source in whooping, if the tube shape and F1 location are 91
36 some detail. To review (and elaborate somewhat): maintained, the timbre becomes increasingly 92
37 close. 93
38 1. When two or more harmonics are below F1, the 8. If H1 is allowed to surpass F1, the voice thins 94
39 timbre is open and ‘chesty’; the more harmonics in timbre and weakens in intensity. Therefore 95
40 below F1, the more open the timbre. when singing above the normal F1 locations, F1 96
41 97
42 98
43 99
44 100
45 101
46 102
47 103
48 104
49 105
50 106
51 107
52 108
53 109
54 110
55 111
56 Figure 3. F1/H1 coupling of whoop or hoot timbre. 112

SLOG_A_679967.indd 3 4/16/2012 1:05:57 PM

 4 K. W. Bozeman
1 •• The ability to let vowels turn over at their nor- 57
2 mal resonant speech locations (an octave below 58
3 their F1 locations). 59
4 •• Vowel integrity with appropriate vowel modifi- 60
5 cation. 61
6 •• Maintenance of chiaroscuro timbre across range 62
7 and transitions. 63
8 64
9 Pedagogic strategies conducive to effective vowel turning 65
10 66
11 •• Encourage a convergent resonator shape, which 67
12 will generate and maintain appropriately low F1 68
13 locations. This can be accomplished by several 69
14 means. The initial means would be the use of a 70
15 noiseless inhalation. A noiseless inhalation 71
16 requires a relatively open throat and convergent 72
17 resonator. However, finding this pre-phonatory 73
18 posture is challenging for students since one’s 74
19 kinesthesia for throat shape is quite misleading. 75
20 Most people perceive /ż/ to be the most open- 76
21 throated posture and /i/ to be the least open- 77
22 throated, when the opposite is in fact the case. 78
Figure 4. Thyro-hyoid space: just below the hyoid bone, laterally
23 and posteriorly.
To circumvent this false kinesthesia, a noiseless 79
24 inhalation can be accomplished by a subtle pal- 80
25 atal lift (incipient yawn) accompanied by cool- 81
26 ing the teeth and front of the mouth upon 82
must be raised in tandem with H1 (8) to main- inhalation, rather than cooling the throat. Wher-
27 83
tain whoop coupling (F1/H1). ever the vocal tract is the coolest it is the nar-
28 84
29 rowest, since air speed will be the quickest and 85
30 the resultant wind-chill effect the strongest there. 86
31 Objectives of passaggio training Shaping the mouth and throat so that the cool- 87
32 The objectives of passaggio training include: ing is in the front of the mouth causes the throat 88
33 to be more open and the resonator 89
34 •• A stable laryngeal position and relatively conver- more convergent. At the same time, shape the 90
35 gent resonator shape for consistency and depth mouth and throat to lower the pitch of any inha- 91
36 of timbre. latory noise. A complementary strategy that 92
37 •• A dynamic laryngeal registration with smooth, encourages a convergent resonator is achieving 93
38 gradual changes for ease, evenness of scale, and a loose, fronted tongue, whose sides may even 94
39 dynamic flexibility (all of which are facilitated make contact with the upper molars. And a third 95
40 by a stable laryngeal position). strategy is to use minimal ‘jawing’ of diction, 96
41 •• A familiarity with the sounds, sensations, and reducing excessive ‘orality’.Vowels do not require 97
42 locations of vowel turning. (Implication: Since exaggerated jaw drop or excessive articulatory 98
43 turning has to do with first formant locations changes in the lower and middle range; rather 99
44 rather than with laryngeal registration, it is erro- vowels can seem fairly close to each other. All of 100
45 neous to think that a voice will turn over on the these strategies improve throat openness and a 101
46 same pitch for all vowels.) convergent resonator, lowering F1. 102
47 103
48 104
49 105
50 106
51 107
52 108
53 109
54 110
55 111
56 Figure 5. Leaping across the pitch of turning. 112

SLOG_A_679967.indd 4 4/16/2012 1:05:58 PM

 Vocal pedagogy  5
1 57
2 58
3 59
4 60
5 61
6 62
7 63
Figure 6. Descending and ascending across the pitch of turning.
8 64
9 65
10 •• Monitor tube length stability. This can be accom- deliberately close in articulation. This is most 66
11 plished by physically monitoring the thyro-hyoid easily accomplished on the close vowels /i/ and 67
12 space in students whose anatomy lends itself to /u/, which turn over well below the passaggio, 68
13 easy identification of laryngeal anatomy (Figure allowing the development of familiarity with the 69
14 4). The thyro-hyoid space should be loose and sounds and sensations of turning over indepen- 70
15 open and not rise or become smaller (as it does in dently of the laryngeal registration challenges 71
16 swallowing) with pitch ascent. Though this space associated with the passaggio. It can then be 72
17 varies in size between individuals, it may also have extended to include the moderately close vowels 73
18 been reduced in size by residual tension from /e/ and /o/. 74
19 unnecessary activation of swallowing muscles dur- •• On ascending leaps, encourage timbral depth, 75
20 ing speech production, a habit that, if entrenched, vowel closeness, and an internalized direction 76
21 should be patiently but persistently countered and and affect. These directives are all intended to 77
22 broken, preferably through therapy with a licensed preserve the vocal tract shape across the turn 78
23 speech language pathologist familiar with appro- and to counteract the instinctive tendency to 79
24 priate therapies, such as massage and stretching of yell, with its shallowing (shortening), opening, 80
25 the thyro-hyoid space, monitoring of the looseness and ‘spreading’ of the resonator. 81
26 and openness of the thyro-hyoid space during •• Glissando up from an open vowel on a low 82
27 speech production, and vocalizing with a low lar- pitch over a large interval to a close vowel on 83
28 ynx and an extended or ‘rolled’ tongue position. a high pitch to identify the kinesthetic and 84
29 Secondly, tube length stability can be assessed by aural poles of openness and closeness 85
30 visually monitoring the location of the Adam’s (Figure 5). 86
31 apple. This method is more useful for males, due •• Repeat this on the same vowel across its pitch 87
32 to their larger average laryngeal anatomy, and for of turning, anticipating, allowing, or if neces- 88
33 those with less fleshy, trimmer necks in which sary, encouraging its passive modification 89
34 structures are more easily visible. Finally, tube (Figure 5). 90
35 length stability can be assessed by aurally monitor- 91
36 ing the consistency of timbral depth/vocal Fach. 92
37 This is less reliable, since it can be falsely mim- Strategies for refining vowel turning 93
38 icked by various articulatory manipulations of 94
39 tongue and lips, but careful listening and observa- •• Model the gradual timbral transition from close 95
40 tion can reduce this liability. to open timbre during scalar descent, followed 96
41 immediately by ascent on the same pitches. The 97
42 The following explorations can assist with the dis- degree of openness/closeness of each ascending 98
43 covery and experience of vowel turning: pitch should match that of the previous descent 99
44 (Figure 6). 100
45 •• Speak an inflective loop on the same vowel •• Finally, ascend through the pitch of turning and 101
46 across its predicted pitch of turning, staying subsequently descend stepwise, allowing the 102
47 103
48 104
49 105
50 106
51 107
52 108
53 109
54 110
55 111
56 Figure 7. Ascending and descending across the pitch of turning. 112

SLOG_A_679967.indd 5 4/16/2012 1:05:58 PM

 6 K. W. Bozeman
1 voice/vowel to close and then reopen across the Declaration of interest: The author reports no 57
2 appropriate pitches (Figure 7). conflicts of interest. 58
3 59
4 60
5 Conclusion References 61
6 1. Sundberg J. Articulatory interpretation of the singing formants.
62
7 Knowledge of the effects of acoustic interactions 63
J Acoust Soc Am. 1974;55:838–44.
8 between source harmonics and the first formant of 2. Titze I, Story B. Acoustic interactions of the voice source 64
9 the vocal tract is crucial to male passaggio pedagogy. with the lower vocal tract. J Acoust Soc Am. 1997;101: 65
10 If tube length and relative closeness/convergence of 2234–43. 66
11 resonator shape are maintained during pitch ascent, 3. Miller DG. Resonance in singing. Princeton, NJ: Inside View
67
the location of the first formant will remain stable, Press; 2008. p. 50–4.
12 4. Bozeman KW. A case for voice science in the voice studio. 68
13 and the second harmonic will surpass the first for- J Sing. 2007;63:267–8. 69
14 mant, appropriately precipitating the timbral shift of 5. Bozeman KW. On the voice: registration strategies for training 70
15 turning over. The variety and predictability of first the male passaggio. Choral J. 2008;48:63–4. 71
16 formant locations, and hence of F1/H2 crossings an 6. Miller DG. Resonance in singing. Princeton, NJ: Inside View
72
octave lower, form a reliable basis both for diagnos- Press; 2008. p. 61–2.
17 7. Miller R. The structure of singing: system and art in vocal 73
18 ing tube length stability and for crafting strategies for technique. New York: Schirmer Books; 1986. p. 116–7. 74
19 training the tube length stability necessary for effec- 8. Sundberg J. The acoustics of the singing voice. Sci Am. 1977; 75
20 tive negotiation of the male passaggio. 236:82–91. 76
21 77
22 78
23 79
24 80
25 81
26 82
27 83
28 84
29 85
30 86
31 87
32 88
33 89
34 90
35 91
36 92
37 93
38 94
39 95
40 96
41 97
42 98
43 99
44 100
45 101
46 102
47 103
48 104
49 105
50 106
51 107
52 108
53 109
54 110
55 111
56 112

SLOG_A_679967.indd 6 4/16/2012 1:05:58 PM