Wavetable synthesis

from Wikipedia, the free encyclopedia
This article or the following section is not adequately provided with supporting documents ( e.g. individual evidence ). Information without sufficient evidence could be removed soon. Please help Wikipedia by researching the information and including good evidence.

As wavetable synthesis (English wavetable = wave table) refers to a technically relatively easy to realize the form of sound generation in synthesizers and sound cards . It was introduced in its original meaning at the end of the 1970s by Wolfgang Palm and represents a special form of playing audio samples . Over time, the term has been used for different types of sample-based sound synthesis.

function

Simple wave synthesis

First of all , a section of the sound that is to be generated later is recorded as a so-called waveform , for example the course of the vibration of an individual tone of a certain pitch, the reference tone , which is electrically detected by a microphone . Most of the time it is the dashed C (c ').

This waveform or part of it can then simply be played back at the original pitch and played back through a loudspeaker . The playback speed is then changed appropriately so that other tones of the scale can be generated.

In order to be able to vary the length of the tone to be reproduced, the measured waveform is usually divided into four areas (see also ADSR ):

  1. The transient part : this is where the sound first builds up and you can hear, for example, the sound of air when you blow a flute or the bow on a violin
  2. a transition part in which the transient part approaches the continuous tone
  3. the continuous tone , i.e. the part that sounds and is repeated until the tone is supposed to fall silent and finally
  4. the swing-out part , where the interrupted note still lingers a little.

By addressing a certain area of ​​the recorded wave, only the appropriate part is played and if necessary. repeated. The quality of the resulting sound depends largely on the skill of the sound designer, who divide the recorded sound into these sections.

Polyphonic sound synthesis

With simple polyphonic synthesis, waveforms of different timbres are structured and stored in a table in the waveform memory of the sound generator. If a polyphonic piece of music is to be played in which different musical instruments occur simultaneously, the waveform memory is read out depending on the desired timbre and the sound of the desired pitch is reproduced either accelerated or slowed down. Often, several samples of different pitches are stored by an instrument in order to reduce the artifacts that result from large differences between the playback and recording frequency.

There are several waveform sections for the creation of one and the same instrument, which are played unmixed and exclusively. The memory contains one or more such waveforms for each voice to be generated.

Classic wave table synthesis

A special form of wavetable synthesis is the technique used by Wolfgang Palm to dynamically switch or cross-fade between several prepared waveforms for each of the voices to be generated while the sound is being generated. For this purpose, several waveforms are stored in a table, which differ in sound to a greater or lesser extent depending on the selection. If the sound generation is started, a waveform is started and - controlled by an LFO or an envelope generator - it is continuously faded to other waveforms. When using an LFO, for example, it oscillates back and forth between waveforms. By passing through the wave field, almost any sound manipulation can take place, depending on which harmonics are contained in the stored waveforms, how many are used and how finely they are resolved. The well-known wave computer 360 from PPG had 64 waveforms per voice.

Advantages and disadvantages

There are significant differences compared to other sound synthesis methods ( subtractive synthesis , additive synthesis , frequency modulation or FM synthesis , physical modeling , granular synthesis , HCM synthesis ). The playback process is technically very easy to implement and therefore widespread even with inexpensive synthesizers or sound cards. It only requires a controller, memory and a digital-to-analog converter .

Because often only a single tone is recorded and different pitches are achieved by accelerating or slowing down the playback of this recording, the tone sounds more and more unnatural as the interval from the reference tone increases , since with real instruments the tone changes slightly with the pitch and this is not shown. To solve this problem, higher quality and more expensive synthesizers, especially stage pianos and high quality electric church organs , use several reference tones that are distributed over the entire pitch spectrum . The same reference tone is then not always modulated during playback, but rather the reference tone that is closest to the pitch to be played. Sometimes each pitch of the instrument to be played is recorded individually and saved as a waveform, which is already comparable to a sampler .

With the help of wavetable synthesis, acceptable results can be achieved within the instrument when simulating real instruments. The use of prepared waves and a number of suitable intermediate versions also enables excessive manipulation of the waveform and thus the generation of abstract sounds that would not be achievable with classic filter technology and this without significant computing power requirements. In modern systems with sufficient computing power, waves are used in a similar way, but mostly modified by means of complex transfer functions in real time and waves from different sources are also faded over at the same time.

See also

Individual evidence

  1. ↑ Uwe G. Hoenig: Wavetable Synthesis - History Ahead of Time. In: hermannseib.com. December 30, 2020, accessed August 15, 2020 .