关于「一个间隙,三个后果:泛音聚焦与鼻窦通量的联合模型」,《泛音双腔、六字诀阻抗与哼鸣一氧化氮》给出的定量结论是什么?

结论:在联合模型中,舌尖齿缝的间隙同时决定泛音聚焦、声源强度与鼻窦通量。聚焦需要很小的间隙,而小间隙通过口内正压削弱声源,使鼻腔声压与一氧化氮增益下降。示例参数下,增益的无约束最优在 0.755 毫米,要求聚焦后最优移到 0.296 毫米,增益从 4.48 降到 3.26。2000 组随机参数中,这一冲突普遍存在,其大小主要由软腭开放程度与声门面积决定。

数据口径:书中的解剖尺寸、压力与浓度多为示意参数;全部计算由书中程序生成;本书不是医疗建议,不宣称任何呼吸法、唱法或发声练习的身体功效。 书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.296, 0.755, 3.26, 4.48

h mmAo mm2Ptg Pasourcepn PaNO gaingap Hzfocus dB
2.0030.00742.70.88559.03.47390.0-9.7
1.0015.00652.10.70480.54.37280.1-12.5
0.609.00553.80.50881.24.40218.4-3.0
0.406.00472.10.34468.93.88178.92.5
0.304.50419.50.23954.63.29155.25.9
0.203.00358.00.11630.92.30126.910.0
0.152.25324.00.04814.01.58110.012.7
0.101.50288.00.0000.01.0089.916.1
0.050.75250.40.0000.01.0063.621.2
unconstrained NO optimum: h = 0.755 mm, gain 4.48
best gap with focus >= 6 dB: h = 0.296 mm, NO gain 3.26,
focus 6.0 dB, transglottal 417.2 Pa
查看完整验算程序
# Chapter 6. p61: One tongue-tip gap h, three consequences. The slit
# (width 15 mm) is the oral exit Ao = w*h of p41; the same Ao, taken as
# the front-tube area against a 300 mm2 back tube, sets the F2-F3 gap
# 4000*atan(sqrt(r))/pi of p11 and so the focus index of p22 (B = 40 Hz,
# drone 150 Hz, harmonic 12). The voice source scales with the excess
# transglottal pressure (Ptg - 300)/(800 - 300); nasal sound pressure =
# 400 Pa * source * nasal share; sinus exchange and first-peak NO gain as
# in p51 to p53 (two sinuses, ostium 2.5 mm, Cs0 = 9 ppm). Illustrative.
import math

RHO, MU, C = 1.14, 1.9e-5, 350.0
PS, AG, AN, PTH, PREF, ABACK = 800.0, 10.0, 6.0, 300.0, 400.0, 300.0
PN, CS0, F0, V = 4.0, 9.0, 150.0, 15e-6
D, LO = 2.5e-3, 6e-3


def g_per_pa(f):
    a = math.pi * D * D / 4
    leff = LO + 1.7 * D / 2
    w = 2 * math.pi * f
    r = (8 * math.pi * MU * LO / (a * a) + leff / (math.pi * (D / 2) ** 3)
         * math.sqrt(2 * RHO * MU * w))
    z = complex(r, w * RHO * leff / a - RHO * C * C / (w * V))
    return 0.5 * f / (abs(z) * w) * 1e6, 2e-5 * a / leff * 1e6


def level(n, fm):
    f, h = n * F0, 1.0
    for fc, bw in fm:
        h *= fc * fc / math.sqrt((fc * fc - f * f) ** 2 + (bw * f) ** 2)
    return 20 * math.log10(h / n)


def focus(gap):
    fm = ((500, 60), (1800 - gap / 2, 40), (1800 + gap / 2, 40),
          (3500, 150))
    return level(12, fm) - max(level(11, fm), level(13, fm))


def joint(h_mm):
    ao = 15.0 * h_mm
    out = ao + AN
    ptg = PS - PS * AG * AG / (AG * AG + out * out)
    src = max(0.0, (ptg - PTH) / (PS - PTH))
    pn = PREF * src * AN / out
    gpp, g0 = g_per_pa(F0)
    gain = (PN + 2 * (gpp * pn + g0) * CS0) / (PN + 2 * g0 * CS0)
    gap = 4000 * math.atan(math.sqrt(ao / ABACK)) / math.pi
    return ao, ptg, src, pn, gain, gap, focus(gap)


print(" h mm  Ao mm2  Ptg Pa  source  pn Pa  NO gain  gap Hz  focus dB")
for h in (2.0, 1.0, 0.6, 0.4, 0.3, 0.2, 0.15, 0.1, 0.05):
    ao, ptg, src, pn, gain, gap, fc = joint(h)
    print(" {:4.2f}  {:6.2f}  {:6.1f}  {:6.3f}  {:5.1f}  {:7.2f}  {:6.1f}"
          "  {:7.1f}".format(h, ao, ptg, src, pn, gain, gap, fc))
best = None
for i in range(1, 2001):
    h = i * 0.001
    ao, ptg, src, pn, gain, gap, fc = joint(h)
    if fc >= 6.0 and (best is None or gain > best[1]):
        best = (h, gain, fc, ptg)
top = max((joint(i * 0.001)[4], i * 0.001) for i in range(1, 2001))
print("unconstrained NO optimum: h = {:.3f} mm, gain {:.2f}".format(
    top[1], top[0]))
print("best gap with focus >= 6 dB: h = {:.3f} mm, NO gain {:.2f},"
      .format(best[0], best[1]))
print("focus {:.1f} dB, transglottal {:.1f} Pa".format(best[2], best[3]))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《双声泛音唱法声道双腔耦合、六字诀气流阻抗谱与呼出气一氧化氮震荡增益》第 6 章「一个间隙,三个后果:泛音聚焦与鼻窦通量的联合模型」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-160 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026160,
  author    = {Fan, Yafu},
  title     = {Two-Cavity Vocal-Tract Coupling in Overtone Singing, Airflow Impedance Spectra of the Liuzijue Six Sounds and Oscillatory Gain in Exhaled Nitric Oxide},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论