Psychoacoustics

from Wikipedia, the free encyclopedia

The psychoacoustics (also psychological acoustics ) is a branch of psychophysics . It deals with the description of the relationship between the human perception of sound as an auditory event and its physical sound field parameters as a sound event . The processing of physical signals into an auditory impression is modeled in several stages. These are assigned to the individual ear and cognitive signal processing . Psychoacoustics examines the relationship between the objective-physical stimulus - the sound waves - and the impression of it in the recipient - such as B. Loudness , sharpness , tonality , roughness , tonality , impulsiveness , fluctuation strength etc. It examines laws in this relationship in order to create hypotheses for processing auditory stimuli and to test them experimentally. Supra-individual or individually different "if-then relationships" between stimulus and psychological experience should be worked out. Important applications of psychoacoustics are in sound effects research, telecommunications , audio data compression and sound design .

parameter

It has proven to be useful to map purely physical parameters such as level , frequency , bandwidth , duration or degree of modulation to aurally appropriate parameters. As a rule, several physical quantities act on one psychoacoustic quantity. This should be able to be assessed as a single sensation independently of other sensations . The scales of psychoacoustic quantities describe the strength of the sensation.

The most common psychoacoustic parameters are the Zwicker parameters loudness (unit sone ), sharpness (unit acum ), tone (unit mel ), roughness (unit asper ) and fluctuation strength (unit vacil ). In addition, tonality and impulsiveness are important variables; they are also used in the formation of rating levels . The phon is the unit of measurement of the psychoacoustic quantity volume level .

Methods

Psychoacoustic tests collect subjective judgments from test subjects. Since these people judge individually, the results of the tests only become valid on the basis of a statistical evaluation of a large number of judgments. The methods for collecting the judgments are divided into classical and adaptive methods, the difference being that the course of an adaptive test is influenced by the test person's judgments, while the classical methods remain unaffected.

Classic methods are: constant stimulus method, size estimation, adjustment method and complete pair comparison.

Adaptive methods are: Forced choice methods (2-AFC, 3-AFC, 4-AFC) and Békésy tracking.

“Curves of equal volume” or more precisely “curves of equal volume level” ( Isophone ) were first created in 1936 by Fletcher-Munson. Measurements by Robinson-Dadson, which were included in the international ISO recommendations R 226 in 1956, also became known. Since 2003 there have been new corrected curves as “Normal equal-loudness-level contours - ISO 226: 2003 Acoustics International Organization for Standardization (ISO) 2nd edition”.

Modeling

How masking effects work

Psychoacoustic tests show that the human hearing can essentially be modeled with a number (e.g. 24 bands according to the Bark scale ) of bandpass filters . This structure is similar to the analysis part of a vocoder .

  • A central term here is the critical bandwidth. If two tones fall in a band, only one tone can be heard, possibly with amplitude modulation or roughness . Only when the frequency spacing of these tones is greater than the critical bandwidth do they fall into two separate filter channels and are therefore perceived as two tones. The critical bandwidth varies across the listening area; it is not constant.
  • The change from rhythm to sound sensation when the frequency of a pulse generator is increased can also be explained by the above model.
  • The perception of the tone ( Mel ) only roughly corresponds to the physically measurable frequency .
  • The perception of the volume only roughly corresponds to the logarithm of the physically measurable sound pressure.

Quiet tones are covered by louder ones that are close by, so they cannot be heard, although they can be physically detected. A loud event that sounds first can obscure a subsequent event. A louder resounding after a quiet event can also obscure the former. This allows conclusions to be drawn about the coupling of the channel data.

  • The transfer of physical measurements to perception is only possible with the greatest care and caution. So are z. B. simple sound level meters are not able to reproduce the impairment caused by noise . Cases are documented in which noise insulation measures were assessed as positive by all test persons, but were classified as deterioration by the simple measuring devices. These discrepancies always occur when the measuring device does not take the above-mentioned hearing method into account.

scientist

Significant works come from:

Various well-known acoustic illusions - comparable to the more well-known optical illusions - illustrate the complexity of hearing.

See also

literature

Individual evidence

  1. ↑ Acoustics226-2003 , sengpielaudio.com (PDF; 30 kB)

Web links