1. A computer-implemented method of handling an audio dialog between a robot and a human user, the method comprising:during said audio dialog, receiving audio data and converting said audio data into text data;
in response to said text data, determining a dialog topic, said dialog topic comprising a dialog content and a dialog voice skin, wherein the dialog content comprises a plurality of sentences, and the dialog voice skin includes audio rendering modifications and voice skin parameters including frequency, velocity and tone;
determining a sentence to be rendered in audio by the robot;
receiving a modification request of said determined dialog sentence, wherein the modification request is a vote of a plurality of parties;
applying one or more moderation rules to the modified determined dialog sentence according to said modification request, wherein the one or more moderation rules comprise one or more filters, said one or more filters comprising blacklists of one or more words or whitelists of one or more words; and
producing a dialog expression based on the modified determined dialog sentence and the one or more moderation rules.