Audio can now be synthesized to accompany haiku. Click 'Play Audio' / 'Mute Audio' to toggle. The sound can only be heard at the moment in browsers that support the WebAudio API. This is a very new API and is not supported in most browsers. This feature works on mainly in Google Chrome.
Audio is produced by one of two methods, depending on if the audio word is used in the haiku.
When the audio word is used, haiku use it to yield an audio sample. Audio is sampled at a much higher rate than visible animation. It currently is computed at x=0, y=0. Audio samples should be in the -1 to 1 range.
When the audio word is not used an older experimental approach is engaged. Audio is produced by sampling the red component at (0, 0.5) to select amplitude, and the red component at (1, 0.5) to select pitch. Both are clipped to [0, 1]. Pitch is mapped to a ten note pentatonic scale. Fractional pitches produce a mix of neighboring tones on the scale. This general scheme was selected to produce non-cacophonous sounds.
This feature is a work in progress and may change.