CAE-USC V2.2 AIMs Speech Synthesis for Restoration

1     Functions

Speech Synthesis for Restoration (CAE-SSR)

Receives	Text List
	Speech Model
Produces	Synthesised Speech

2      Reference Model

Figure 1 depicts the Reference Model of the Speech Synthesis for Restoration (CAE-SSR) AIM

Figure 1 – Reference Model of the Speech Synthesis for Restoration (CAE-SSR) AIM

3      Input/Output Data

Input data	Semantics
Text List	List of Text items to be converted into Speech.
Neural Network Speech Model	A Neural Network Model trained on Speech Segments for Modelling and used to synthesise replacements for the entire Damaged Segment or Damaged Sections within it.
Output data	Semantics
Synthesised Speech	Speech synthesised by Neural Network Speech Model.

4      SubAIMs

No SubAIMs.

5      JSON Metadata

https://schemas.mpai.community/CAE1/V2.2/AIMs/SpeechSynthesisForRestoration.json

6     Profiles

No Profiles