1   Definition

A Data Type able to generate speech according to a specified set of features.

2   Functional Requirements

The generated speech can be perceived as:

  1. Having been generated by a specific human.
  2. Have a specific language or dialect intonation.
  3. Not have any specific connotation.


It can be implemented as a Neural Network trained to generate utterances with specific Speech Descriptors.

3   Syntax

No syntax provided.

4   Semantics

No Semantics provided.

5   Data Formats

No entry here.

6   To Respondents

Respondents are requested to:

  1. Comment on the characterisation of Speech Model.
  2. Propose Speech Model Formats.