Figure 2.
The person appears with a thought bubble containing an explosion symbol and a waveform near the mouth. Three pathways are labelled Imagine, Represent and Synthesis. The upper two pathways begin with Imagine and proceed to Represent, then towards Synthesis. A lower pathway begins from the vocalised waveform, proceeds through Imagine and Represent, and then towards Synthesis. The synthesis pathways connect to a network labelled PronounSE. An arrow from the synthesis stage points towards Desired Sound.Sound creation flow with PronounSE. This enables a human-in-the-loop sound design workflow, in which users iteratively perform mimicking and synthesizing audio to gradually approach the desired sound