<-AI Workflows Go to ToC Data Types->

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

1. Technical Specifications

Table 1 provides the links to the specifications and the JSON syntax of all AIMs specified by Technical Specification: Context-based Audio Enhancement (MPAI-CAE) – Use Cases (CAE-USC) V2.3. The AI Modules specified by CAE-USC V2.3 supersede those specified by previous versions. These may be used if their Version is explicitly signaled. AIMs in bold are Composite.

Table 1 – Specifications and JSON syntax of AIMs used by CAE-USC V2.3

Acronym AIM Name JSON Acronym AIM Name JSON
CAE-AAP Audio Analysis for Preservation File CAE-PAP Packaging for Audio Preservation File
CAE-AAT Audio Analysis Transform File CAE-PEI Prosodic Emotion Insertion File
CAE-ABS Audio Basic Scene Description File CAE-SDS Speech Detection and Separation File
CAE-ADP Audio Description Packaging File CAE-SF1 Speech Feature Analysis 1 File
CAE-AMX Audio Descriptors Multiplexing File CAE-SF2 Speech Feature Analysis 2 File
CAE-AOI Audio Object Identification File CAE-SFD Sound Field Description File
CAE-ASD Audio Scene Description File CAE-SMC Speech Model Creation File
CAE-ASE Audio Separation and Enhancement File CAE-SRA Speech Restoration Assembly File
CAE-ASL Audio Source Localisation File CAE-SSR Speech Synthesis for Restoration File
CAE-AST Audio Synthesis Transform File CAE-TAR Tape Audio Restoration File
CAE-EFP Emotion Feature Production File CAE-TIC Tape Irregularity Classification File
CAE-NCM Noise Cancellation Module File CAE-VAP Video Analysis for Preservation File
CAE-NEI Neural Emotion Insertion File

2. Conformance Testing

An implementation of an AI Module conforms with CAE-USC V2.3 if it accepts as input and produces as output Data and/or Data Objects conforming with those specified by CAE-USC V2.3. Note that a Data Object is defined as the combination of Data of a Data Type and its Qualifier.

The Conformance of an instance of Data of a Data Type is to be expressed by a sentence like “Data validates against the Data Type Schema”. This means that:

  • Any Data Sub-Type is as signaled in the Qualifier.
  • The Data, File, and/or Stream have the Format signaled by the Qualifier.
  • The Attributes of the Data have the type or validate against the Schemas specified in the Qualifier.

The method to Test the Conformance of a Data or Data Object instance is specified in the Data Types chapter.

3. Performance Assessment

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

  1. Quality: Performance Assessment measures how well an AIM performs its function, using a metric that depends on the nature of the function, e.g., the perceived quality of a stereo audio.
  2. Bias: Performance Assessment measures how well an AIM performs its function, using a metric that depends on a bias related to certain attributes of the AIM. For instance, an Audio Source Localisation AIM tends to be more accurate in localising certain types of Audio Objects that with Objects of another type.
  3. Legal compliance: Performance Assessment measures how well an AIM performs its function, using a metric that assesses its accordance with a certain legal standard.
  4. Ethical compliance: the Performance Assessment of an AIW can measure the compliance of an AIW to a target ethical standard.

The current USC-CAE V2.3 Standard does not provide AIW Performance Assessment methods.

Assessing the Performance of an AIM may be complex because of the multiple dimensions involved with the input and output data of an AIM.

<-AI Workflows Go to ToC Data Types->