Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

Share this Page URL

5.7 Summary 223 matching models and stages from several sources. Imagine having an "energy-generation" module of a violin fed into the "resonance and damping" section of a drum and then sent to an "amplifier" stage of a trumpet. The resulting sonority will be something totally new yet extremely musical and inspiring! 5.6.9 granular Synthesis This type of synthesis is newer than some of the more established and widely used techniques analyzed so far. Even though the concept on which granular synthesis (GS) is based can be traced back to the end of the nineteenth century, actual practical use in synthesis was developed only around the late 1940s and early 1950s. The main idea behind GS is that a sound can be synthesized starting from extremely small acoustic events called grains, which are usually between 10 and 100 ms in length. Granular synthesiz- ers start from a waveform (it can be any type of waveform, such as FM, wavetable, or sampled) that is "sliced" to create a "cloud", which is made up of hundreds or even thousands of grains. A completely new and complex waveform is then assembled by reordering the grains according to several algorithms that can be randomly based or functional. One advantage of GS is that the sounds created from such small units can constantly change and evolve, depending on the grain-selection algorithm. Since the grains forming the sound cloud are so small and so numerous, their repetition is perceived not as a loop (as in the case of wavetable synthesis or sam- pling) but instead as "quantum" material that constitutes the basic substance of which sounds are made. Grains and the resulting waveforms can be processed using more standard modifiers, such as enve-