Embodied negotiation agent and platform

A method comprising:

• obtaining, at a software agent, a representation of human speech signals uttered within an environment including an avatar representing said software agent, said human speech signals having been directed to said avatar;
• obtaining at least one non-speech behavioral trace from within said environment, said trace overlapping temporally with said human speech signals and being representative of non-speech behavior directed to said avatar;
• interpreting said human speech signals and said at least one non-speech behavioral trace, with said software agent, as a first act with associated parameters;
• generating a proposed act responsive to said first act;
• and
• submitting said proposed act to an environment orchestrator.

A system comprising:

• a memory;
• a non-transitory computer readable medium comprising computer executable instructions;
• and
• at least one processor, coupled to said memory and said non-transitory computer readable medium, and operative to execute said instructions to be operative to:
• transcribe human speech signals uttered within an environment including one or more avatars representing one or more software agents, said human speech signals being directed to at least one of said one or more avatars;
• obtain at least one non-speech behavioral trace within said environment, said trace being representative of non-speech behavior directed to said at least one of said one or more avatars;
• forward said transcribed human speech signals and said at least one non-speech behavioral trace to said one or more software agents;
• obtain, from at least one of said one or more software agents, a proposed act;
• and
• responsive to said obtaining, issue a command to cause one of said one or more avatars corresponding to said at least one of said one or more software agents from which said proposed act is obtained to emit synthesized speech and to act visually in accordance with said proposed act.