Basic architecture of D2S
The picture above shows the basic architecture of D2S. The system consists of two main components: the Language Generation Module (LGM) and the Speech Generation Module (SGM).
The LGM takes data as input and generates a natural language text expressing these data. The Prosody module which is part of the LGM annotates this text with prosodic markers indicating the placement of accents and phrase boundaries. This information is used by the SGM, which turns the generated text into a speech signal.