Layered Examples

Audio example 11 was created by layering many transposed copies of a single shape. The score for this example is printed in table 5.9.

Table 5.9: The score for audio example 11.

The structure for this example has 3 points; the first and the last points both have the ``silent'' option on; therefore, it is only the middle point ``a2'' which creates any sound. The ``window'' used for this example is the final 0.3 seconds of the spoken word ``light'' without the letter `l'. Therefore, the ``window'' starts with a voiced sound and ends with a noisy fricative, and this structure is magnified to 20 seconds in the duration of the example. Notice the use of the ``interpol'' option. By default, the system does not interpolate any of the values either when applying amplitude windows or when looking up tables. The option ``finterpol'' means to interpolate during a table look up, and ``ainterpol'' to interpolate when applying amplitude windows, and ``interpol'' to interpolate in both cases. The spectrogram for this example can be seen in figure 5-5. This sound was used as the opening sound of Morphosis (1992), which is a piece composed by the author using this system and is partly described in appendix C.

Figure 5-5: The spectrogram of sound example 11 is illustrated.

Audio example 12 has been created by applying the same type of procedure to a longer melody of a cello sound.

The score for audio example 13 is printed in Table 5.10. This example adds many layers of looped sound of a piano note. The entry ``table: "piano/d2":25000-157300;'' picks 3 seconds of a sampled piano sound. The numbers specified in the table entry are sample numbers, and this option is provided for precise definition of tables. The point ``a2'' is ``silent''. The point ``a1'' segments the time by a factor of 0.95 while the frequency factor of it is 1.052632 which is $1/0.95$. Thus, as the segments get shorter the frequency value gets larger by the same factor. In this way, every layer becomes 20 ($60 / 3$) notes. This example also shows how we can create stereo outputs. The number of channels are specified in the ``snd'' object by the ``nchnls: 2;'' entry. The factors ``ch1'' and ``ch2'' in the point ``a1'' are applied to channel 1 and channel 2 respectively. The factors for ``ch1'' and ``ch2'' are specified as expressions by using the ``if'' function. Three arguments are passed to ``if''; the first is a condition, and the value of the ``if'' function is either the second or the third argument depending on the truth value of the first argument. Therefore, in this case the values of both ``ch1'' and ``ch2'' are 0.5 if we are in the first level (rec_level == 1); otherwise their values is either 1 or 0 depending on the level. Thus, except for the first level, every other level of the sound is assigned to either channel one or two. A similar version of this sound was used for the ending of Morphosis.

Table 5.10: The score for audio example 13.