5D Library › Turning the Hearing Inward · 中文
Chapter 3. Stage One: Acoustic Decoupling
“Turn around your inverted faculty of hearing.” (Mañjuśrī’s verse)
The first stage does not ask the singer to do anything new. It asks them to stop doing one thing: trying to make the voice they hear inside their head match a voice they imagine the audience hears.
The problem: two voices, one judge
Chapter 2 showed that a singer hears a blend of two signals, the air-conducted voice that everyone hears and a bone-conducted voice that only the singer hears. The bone-conducted part is strongest between about 1 and 2 kHz, changes from vowel to vowel, and reaches the singer louder and earlier than any listener in the room.
Most singers know, at least vaguely, that they sound different on a recording. Far fewer draw the practical conclusion: the sound in your head is not a reliable guide to the sound in the room. When a singer judges the result by the inner sound, two mistakes are common.
- Chasing size. The inner voice sounds small or dull, so the singer adds effort to make it “bigger”. The audience hears mostly the extra effort.
- Chasing the recording. The singer has learned what they “should” sound like from recordings and tries to reproduce that sound in their own head. But the recording was made from air conduction alone; no amount of effort makes the inner blend sound like it.
Both mistakes have the same physical outcome: unnecessary effort in the throat.
Calculation
Why your own voice always sounds loud to you. In open space, sound pressure level falls by about 6 dB each time the distance doubles (the inverse-square law), so between a near and a far point the level difference is The singer’s ear is about from the mouth; a listener might be away:
import math
r_ear, r_listener = 0.15, 10.0
dL = 20 * math.log10(r_listener / r_ear)
print(f"free-field difference: {dL:.1f} dB")
# free-field difference: 36.5 dB
# What extra effort buys: level change for a given factor of acoustic power
for factor in (2, 4, 10):
print(f"x{factor} power -> +{10*math.log10(factor):.1f} dB")
# x2 power -> +3.0 dB
# x4 power -> +6.0 dB
# x10 power -> +10.0 dBIn a real room, reflections narrow the 36.5 dB gap, but the singer still hears themselves far louder than anyone else does, and through a channel (bone) that no listener shares. Doubling the acoustic power adds only 3 dB; a common rule of thumb is that a sound must rise by about 10 dB, ten times the power, to be heard as roughly twice as loud. Pushing is an expensive way to be heard.
Noise makes us push without deciding to
Evidence
In 1911 the French laryngologist Étienne Lombard reported that people raise their voices in noise without meaning to. The Lombard effect has since been documented in many species; Henrik Brumm and Sue Anne Zollinger’s review of a century of research describes it as an automatic adjustment that is hard to suppress by intention alone (Brumm and Zollinger 2011). On stage, with a band, a monitor mix and an audience, a singer is in exactly the conditions that produce it.
The point for the first stage is that effort is often not chosen. It is triggered: by noise, by the wish to sound bigger, by nerves. A practice that targets effort has to work with attention, because the effort itself does not wait for permission.
Effort lands in the muscles around the larynx
Evidence
The larynx hangs in the neck from a sling of muscles above it (suprahyoid) and below it (infrahyoid). These extrinsic laryngeal muscles position and steady the larynx; they do not produce the voice. Excess activity in them is a common finding in muscle tension dysphonia, a voice disorder without structural damage to the vocal folds, and clinical reviews describe raised larynx position, neck tension and effortful voice as typical signs (Van Houtte et al. 2011).
Note
This book does not diagnose or treat voice disorders. Hoarseness lasting more than a few weeks, pain, or a change in the voice without obvious cause needs a doctor’s examination. The exercises below are listening exercises for healthy singers. They involve light touch only; never press on the neck.
The exercises of Stage One
Hypothesis · not established · Stage One: acoustic decoupling (the author’s method)
The aim of the first stage is to separate the two routes in experience, so that the singer knows from the inside which part of what they hear belongs only to them, and to remove the effort spent chasing it. The author’s hypothesis is that a singer who has separated the two routes will push less and judge pitch more steadily.
1. Two-route listening (3 minutes). Hum on a comfortable note with the ears open. Then close the ear canals gently with the fingertips and hum the same note at the same volume: the hum becomes louder and darker. This is mostly the bone route (Chapter 2). Open and close a few times. Then do the same on the vowels /a/, /i/ and /u/, and notice which vowel changes most when the ears are closed. Do not try to change the sound; only notice.
2. Record and compare (3 minutes). Place a phone about a metre in front of you and record one short phrase sung comfortably. Listen back once. Name three differences between the recording and what you heard while singing (for example: “brighter”, “less bass”, “more breath”). Do not sing the phrase again to fix them. The purpose is to learn the size and direction of the difference, not to close it.
3. The light-touch check (4 minutes). Rest the fingertips of one hand lightly on the front of the neck, just above the Adam’s apple, and the other hand’s fingertips at the side of the neck. Sing a slow five-note scale up and down at a comfortable volume. If the area under the fingers hardens, pushes outward or rises sharply, repeat at a lower volume and with less effort until the scale passes with little change under the fingers. This is the singer’s working level for the rest of the session.
4. The one rule. For the rest of practice, the singer gives up one goal: making the inner sound match the recording or any other outer model. The recording is information about the room. The inner sound is information about the body. They are not supposed to be the same.
What the first stage does not claim
The first stage does not claim that bone-conducted sound is “truer” or more spiritual than air-conducted sound, nor that the outer sound does not matter. The audience hears the outer sound, and it matters very much. The claim is narrower: that a singer who keeps comparing two signals that can never match will keep adding effort, and that the first step toward a freer voice is to stop the comparison.
The sutra’s phrase for this is “turn around your inverted faculty of hearing”. Ordinary hearing faces outward, toward the sound as an object to be judged. The first stage does not yet turn it fully around; it only notices which way it is facing.
Summary
- A singer hears a blend of air- and bone-conducted sound that no listener hears; the inner sound is not a reliable guide to the sound in the room.
- The singer hears themselves tens of decibels louder than a distant listener does, and extra power buys little extra loudness.
- Noise and the wish to sound bigger trigger effort automatically (the Lombard effect); the effort lands in the muscles around the larynx.
- Stage One separates the two routes in experience and replaces the goal “make it sound like the recording” with “notice the difference and leave it”.
- The claim that this reduces pushing and steadies pitch is a hypothesis.
References
Fan, Y. (2026). Turning the Hearing Inward: A Bioacoustic Study of the Ergen Yuantong Listening Method [反闻闻自性] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23666557 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).