5D Library › Singer Formant Clustering, Glottal Closure Quotients and Acoustic Authority Psychophysics · 中文
Chapter 4. Final Falls and Strategic Pauses: Measuring Two Prosodic Signals
The final fall of a statement
A statement in English and in Mandarin usually ends with a fall in pitch; a final rise (“uptalk”) is often heard as uncertainty. The slope of the final fall is easy to state and hard to measure, because pitch trackers make octave errors: in a creaky or breathy ending a tracker may report a frequency twice or half the true one. p07 measures the last 300 ms of synthetic falls, level endings and rises with two estimators.
Calculation
# Chapter 4. p07: Measuring the final fall of a statement. The last 300
# ms of an utterance in 10 ms frames, pitch in semitones: a fall of 6
# semitones (-20 st/s), a level ending, or a rise of 4 (uptalk). The
# pitch tracker adds noise of sd 0.5 st per frame and, with chance q
# per frame, an octave error of +12 or -12 st. The slope is estimated
# by least squares (OLS) or by the median of pairwise slopes
# (Theil-Sen) and called "fall" below -5 st/s, "rise" above +5, else
# "level". 2,000 utterances per case. Seed 99.
import random
T = [i * 0.01 for i in range(30)]
SHAPES = (("fall", -20.0), ("level", 0.0), ("rise", 4 / 0.29))
def ols(ys):
mt, my = sum(T) / 30, sum(ys) / 30
return sum((t - mt) * (y - my) for t, y in zip(T, ys)) / sum(
(t - mt) ** 2 for t in T)
def theil_sen(ys):
s = sorted((ys[j] - ys[i]) / (T[j] - T[i])
for i in range(30) for j in range(i + 1, 30))
return s[len(s) // 2]
def label(b):
return "fall" if b < -5 else "rise" if b > 5 else "level"
if __name__ == "__main__":
rng = random.Random(99)
print(" octave errors OLS wrong Theil-Sen wrong (share of calls)")
for q in (0.0, 0.02, 0.05, 0.10):
bad = [0, 0]
for _ in range(2000):
name, slope = SHAPES[rng.randrange(3)]
ys = []
for t in T:
y = slope * t + rng.gauss(0, 0.5)
if rng.random() < q:
y += rng.choice((12, -12))
ys.append(y)
bad[0] += label(ols(ys)) != name
bad[1] += label(theil_sen(ys)) != name
print(" {:13.2f} {:9.3f} {:15.3f}".format(
q, bad[0] / 2000, bad[1] / 2000))
# octave errors OLS wrong Theil-Sen wrong (share of calls)
# 0.00 0.000 0.000
# 0.02 0.076 0.000
# 0.05 0.152 0.000
# 0.10 0.227 0.002Without octave errors both estimators call every ending correctly. With an octave error in 0.02 of frames, ordinary least squares mislabels 0.076 of endings; at 0.05, 0.152; at 0.10, 0.227. The median of pairwise slopes (Theil-Sen) is almost unaffected: 0.002 at 0.10. Any rule that rates a speaker by the steepness of the final fall should use a robust slope, or the rating will mostly measure the tracker.
What counts as a pause
The “strategic pause ratio” of a persuasive speaker is often given as 18% to 25% of speaking time. p08 shows that such a number cannot be read without its definition: what level counts as silence, and how long a silence must be to count as a pause.
Calculation
# Chapter 4. p08: How a "pause ratio" depends on its definition. Sixty
# seconds of synthetic speech in 10 ms frames: phrases of 1.5 to 4 s
# separated by pauses (log-normal, median 450 ms); inside phrases,
# syllables at 5 a second, each followed with chance 0.3 by a stop
# closure of 40 to 120 ms. Speech frames have levels around 65 dB (sd
# 6); silent frames sit at a 35 dB noise floor (sd 3). A frame is
# silent if its level is below a threshold; the pause ratio is the
# share of time in silent runs at least as long as a minimum. 200
# recordings, seed 99.
import random
from math import exp
def recording(rng):
frames = []
while len(frames) < 6000:
end = len(frames) + rng.randint(150, 400)
while len(frames) < end:
frames += [True] * rng.randint(12, 28)
if rng.random() < 0.3:
frames += [False] * rng.randint(4, 12)
frames += [False] * max(1, round(45 * exp(rng.gauss(0, 0.5))))
return [rng.gauss(65, 6) if s else rng.gauss(35, 3)
for s in frames[:6000]]
def ratio(levels, thr, min_ms):
total, run = 0, 0
for v in levels + [99.0]:
if v < thr:
run += 1
else:
if run * 10 >= min_ms:
total += run
run = 0
return total / len(levels)
if __name__ == "__main__":
rng = random.Random(99)
recs = [recording(rng) for _ in range(200)]
mins = (0, 100, 150, 250, 400)
print(" threshold (dB) minimum pause (ms):")
print(" " + " " * 16 + "".join(" {:>8d}".format(m) for m in mins))
for thr in (45, 50, 55):
row = [sum(ratio(r, thr, m) for r in recs) / 200 for m in mins]
print(" {:14d} ".format(thr) + "".join(
" {:8.3f}".format(x) for x in row))
# threshold (dB) minimum pause (ms):
# 0 100 150 250 400
# 45 0.241 0.195 0.157 0.152 0.127
# 50 0.245 0.196 0.157 0.153 0.128
# 55 0.277 0.197 0.157 0.153 0.129The same recordings have a pause ratio of 0.241 if every silent frame counts, 0.195 if silences shorter than 100 ms (mostly stop closures inside words) are excluded, 0.157 at 150 ms, 0.152 at 250 ms and 0.127 at 400 ms. The silence threshold matters less here (0.245 at 50 dB and 0.277 at 55 dB without a minimum duration), but only because the synthetic noise floor is far below speech. A “pause ratio of 18% to 25%” can describe the same speaker as 0.127 or as 0.241 depending on the definition; a band only makes sense with the minimum pause duration attached.
Hypothesis · not established
The author’s interpretation (not yet tested): when pauses are defined as silences of at least 250 ms, speakers rated most persuasive in public speeches have pause ratios in a band of width no more than 0.10, and the centre of that band differs between languages. What would show it wrong: in a multi-language corpus of rated speeches, the pause ratios of the top-rated tenth of speakers span more than 0.10, or their medians do not differ between languages.
Summary
A robust slope estimator is needed to measure final falls once pitch trackers make octave errors. A pause ratio depends mostly on the minimum pause duration: the same recordings gave 0.241 or 0.127, so a target band without a definition is not a measurement.
Fan, Y. (2026). Singer Formant Clustering, Glottal Closure Quotients and Acoustic Authority Psychophysics: Subglottal Pressure Dynamics, Pitch-Floor Stability and Pause-to-Phoneme Persuasion Ratios [歌手共振峰簇、声门闭合商与声学权威度心理物理学] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090258 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).