<- Use Cases    Go t o ToC     Data Formats- >

AI Modules (AIM) are interconnected processing units composing an AI Workflow (AW). There are two types of AIM:

  1. Composite AIMs are AI Modules composed of multiple AI Modules.
  2. Basic AIMs do not include (or do not expose) AIMs inside.

This Chapter specifies the Composite AIMs of Table 1 .

Table 1 – Composite AIMs of MPAI MMC

Acronym AIM Name Versions
MMC-PSE Personal Status Extraction V2.1
MMC-TST Text and Speech Translation V2.1

This Chapter also specifies the Basic AIMs of Table 2.

Table 2 – Basic AIMs of MPAI MMC- Note that V12.2 is currently not approved.

Acronym AIM Name Versions
MMC-ASR Automatic Speech Recognition V2.1 V2.2
MMC-EDP Entity Dialogue Processing V2.1
MMC-ISD Input Speech Description V2.1
MMC-ITD Input Text Description V2.1
MMC-NLU Natural Language Understanding V2.1
MMC-PMX Personal Status Multiplexing V2.1
MMC-PSE Personal Status Extraction V2.1
MMC-PSI PS-Speech Interpretation V2.1
MMC-PTI PS-Text Interpretation V2.1
MMC-SCM Summary Creation Module V2.1
MMC-SIR Speaker Identity Recognition V2.2
MMC-SIR Speaker Identity Recognition V2.1
MMC-TST Text and Speech Translation V2.1
MMC-TTS Text-to-Speech V2.1
MMC-TTT Text-to-Text Translation V2.1
MMC-TVD TV Diarisation V2.2

 

 

Acronym AIM Name Versions
MMC-ASR Automatic Speech Recognition V2.1 V2.2
MMC-BST Bidirectional Speech Translation V2.1
MMC-CAS Conversation About a Scene V2.1
MMC-CPS Conversation with Personal Status V2.1
MMC-CWE Conversation with Emotion V2.1
MMC-HCI Human-CAV Interaction V2.1
MMC-MQA Multimodal Question Answering V2.1
MMC-MST One-to-Many Speech Translation V2.1
MMC-SIR Speaker Identity Recognition V2.2
MMC-TVD TV Diarisation V2.2
MMC-UST Unidirectional Speech Translation V2.1
MMC-VMS Virtual Meeting Secretary V2.1

 

<- Use Cases    Go t o ToC     Data Formats- >