ECHO is a smart mirror: you sit in a booth and see your reflection. Talk to the other you and your reflection will answer by repeating sentences and words you said before. The device’s AI ensures that the tone is harmonious within the conversation, with an often-enigmatic meaning.
The prosodic aspects are called the « musical elements » of language. ECHO pays attention to the prosodic “attitudes” that it deduces from the analysis of the ongoing conversation. It is particularly effective at this task, as it is not influenced by the meaning of words. Because of this uniquely musical and asemantic approach, the machine can adapt to any language.
This prosodic analysis mediates ECHO’s relationship with the visitor. As music is governed by both the laws of mathematics and those of emotion, the link between the machine’s predictive system and the human is fluid and natural.
The system records the words of the discussion, detects predictable situations in terms of prosody, and then repeats them at the appropriate moment. This disruption can create surprise, embarrassment, and laughter.
According to Bergson, what is required to transform all the seriousness and the gravity of our life into a comedy is “merely to fancy that our seeming freedom conceals the strings of a dancing-Jack, and that we are, as the poet says :
… humble marionettes
The wires of which are pulled by Fate.”