Sharpness is a psychoacoustic attribute that describes the proportion of high-frequency energy
in a sound. Sounds dominated by high-frequency content are perceived as sharp, piercing, or
bright, while those dominated by low-frequency content are perceived as dull or dark. Sharpness
is a key contributor to annoyance and sound quality in consumer products, vehicles, and HVAC
systems.
Standard
Standard
Method
Year
DIN 45692
Measurement technique for the simulation of the auditory sensation of sharpness
2009
DIN 45692 defines the calculation procedure based on the specific loudness spectrum. It
supersedes earlier proposals by Bismarck (1974) and Aures (1985) by specifying a standardized
weighting function and normalization.
Theory
Relationship to Loudness
Sharpness is derived from the specific loudness spectrum N'(z) computed according to
ISO 532-1 or DIN 45631. The specific loudness values are weighted by a function g(z) that
emphasizes contributions above approximately 15.8 Bark (corresponding to roughly 3 kHz).
Weighting Function
The weighting function g(z) is defined as:
g(z) = 1 for z ≤ 15.8 Bark
g(z) = 0.066 · e0.171 z for z > 15.8 Bark
Sharpness Calculation
Total sharpness is obtained by computing the weighted first moment of the specific loudness
distribution, normalized by total loudness:
S = k · ∫024 Bark N'(z) · g(z) · z · dz / N [acum]
The constant k = 0.11 is chosen so that a narrow-band noise centered at 1 kHz with a level
of 60 dB and a bandwidth of one critical band yields exactly 1 acum.
Parameters
Parameter
Options
Description
Loudness Standard
ISO 532-1 / DIN 45631/A1
Underlying loudness model for specific loudness computation
Field Type
Free field / Diffuse field
Ear transfer function applied in the loudness calculation
Weighting Function
DIN 45692 (default) / Aures / Bismarck
High-frequency weighting variant
Signal Type
Stationary / Time-varying
Determines whether running analysis is performed
Output Quantities
Quantity
Unit
Description
S
acum
Overall sharpness value
S(t)
acum
Sharpness vs Time trace for time-varying signals
Sharpness vs Time
In time-varying mode, sharpness is computed at each time step using the instantaneous
specific loudness spectrum. The resulting S(t) trace reveals how high-frequency content
evolves over time, which is particularly useful for:
Engine run-up analysis where intake or turbo whine becomes prominent at higher RPM
Transient events such as switch clicks or relay activations
HVAC blower noise as fan speed ramps up
Interpretation Guidelines
Reference value: 1 acum is defined as the sharpness of a narrow-band noise
one critical band wide, centered at 1 kHz, at a level of 60 dB.
< 1.0 acum — Low sharpness; warm, dull character (e.g., low-frequency hum).
1.0 – 2.0 acum — Moderate sharpness; typical for broadband noise with balanced spectrum.
2.0 – 3.5 acum — Elevated sharpness; noticeable high-frequency emphasis.
> 3.5 acum — High sharpness; piercing, potentially annoying sound character.
Practical Application Notes
Sharpness is strongly correlated with subjective annoyance. In many sound quality models,
sharpness enters with a high weighting coefficient alongside loudness.
Reducing sharpness in a product sound can often be achieved by adding low-frequency damping
material or modifying fan blade geometry to shift spectral energy downward.
When using Aures' weighting function instead of DIN 45692, sharpness values include a
loudness-level dependency; results are generally higher for low-level sounds.
Always ensure the underlying loudness calculation uses the correct field type, as the
specific loudness spectrum directly determines sharpness.
Standard References
DIN 45692:2009 — Measurement technique for the simulation of the auditory sensation of sharpness
G. von Bismarck, "Sharpness as an attribute of the timbre of steady sounds," Acustica, vol. 30, 1974
W. Aures, "Berechnungsverfahren für den sensorischen Wohlklang," Acustica, vol. 59, 1985
E. Zwicker & H. Fastl, Psychoacoustics: Facts and Models, 3rd ed., Springer, 2007