<-AI Modules     Go to ToC        Datasets->

1. Technical Specification 2. Conformance Testing 3. Performance Assessment

1. Technical Specifications

This page gives the links to the specification of Data Types specified by Technical Specification: Multimodal Conversation (MPAI-MMC) V2.3. All previously specified MPAI-MMC Data Types are superseded by those specified by V2.3.

Cognitive State Emotion Face Personal Status Gesture Personal Status
Intention Meaning Personal Status Social Attitude
Speech Basic Scene Descriptors Speech Basic Scene Geometry Speech Descriptors Speech Object
Speech Overlap Speech Personal Status Speech Scene Descriptors Speech Scene Geometry
Summary Text Descriptors Text Object Text Personal Status
Text Segment Text Word

2     Conformance testing

A Data instance of a Data Type specified by MPAI-MMC V2.3 Conforms with it if the JSON Data validate against the relevant MPAI-MMC V2.3 JSON Schema and if the Data Conforms with the relevant Data Qualifier, if present. MPAI-MMC V2.3 does not provide method for testing the Conformance of the Semantics of the Data instance to the MPAI-MMC V2.3 specification.

Conformance testing can be performed by a human using a JSON Validator to verify the Conformance of the syntax of JSON Data to the relevant JSON Schema; and, if the Data has a Qualifier, to verify that the syntax of the Data conforms with the relevant values in the Data Qualifier. Alternatively, Conformance testing can be performed by software implementing the steps above.

3     Performance Assessment

Performance Assessment provides methods of assessing the performance of an Data instance. Performance may have various connotations, such as:

  1. Quality: Performance Assessment measures the quality of the Data instance using a metric that depends on the nature of the Data, e.g., the word error rate (WER) of a string of characters representing a sentence compared to an idea sentence.
  2. Bias: Performance Assessment uses a metric that depends on the bias in the Data compared with reference Data related to certain attributes of the Data.  For instance, the Data may contain information about a particular geographic area when the ideal data do not .
  3. Legal compliance: Performance Assessment uses an appropriate metric to measure how well the Data instance complies with with a certain legal standard.

<-AI Modules     Go to ToC        Datasets->