← Registry / Multimodal Conversation
AIM · composite
Affective Multimodal Conversation
MMC-AMC-V2.5
Submissions for this AIM must match this definition: the same ports, API profile and internal wiring.
Description
Affective Multimodal Conversation: speech-only affective dialogue. Automatic Speech Recognition recognises the speech; Natural Language Understanding produces the Text Descriptors, the (refined) NLU Text and the Text Personal Status; Personal Status Extraction reads the Speech Personal Status from the Speech Object and multiplexes it with the Text Personal Status into the entity's Personal Status; Entity Dialogue Processing produces the machine's response Text and Personal Status, aware of both the meaning and the feeling; Response and Scene Rendering renders the Portable Avatar. The Summary of the conversation is internal to Entity Dialogue Processing. Acquisition and delivery are the User Agent, not sub-AIMs.
Inputs and outputs 6 in · 2 out
Inputs
- InputSpeech optional OSD-BSO-V1.5
- InputFace optional OSD-BVO-V1.5
- InputText optional OSD-BTO-V1.5
- AvatarModel OSD-B3O-V1.5
- MachineID optional string
- Summary optional MMC-SUM-V2.5
Outputs
- PortableAvatar PAF-PAV-V1.6
- EditedSummary MMC-SUM-V2.5
Built from 5 AIMs
| AIM | Code | Instances |
|---|---|---|
| Automatic Speech Recognition | MMC-ASR-V2.5 | 1 |
| Natural Language Understanding | MMC-NLU-V2.5 | 1 |
| Personal Status Extraction | MMC-PSE-V2.5 | 1 |
| Entity Dialogue Processing | MMC-EDP-V2.5 | 1 |
| Response and Scene Rendering | PAF-RSR-V1.6 | 1 |
Documentation
Implementations 0
None published yet.