MaryTTS: The MARY Text-to-Speech System

Project Website:
Github Link: None

Description

MaryTTS is an open-source, multilingual Synthesis platform written in Java. It was originally developed as a collaborative project of DFKI’s Language Technology Lab and the Institute of Phonetics at Saarland University. It is now maintained by the Multimodal Speech Processing Group in the Cluster of Excellence MMCI and DFKI.

As of version 5.2-SNAPSHOT, MaryTTS supports German, British and American English, French, Italian, Swedish, Russian, Turkish, and Telugu; more languages are in preparation. MaryTTS comes with toolkits for quickly adding support for new languages and for building unit selection and HMM-based synthesis voices.

Four parts of the system can be distinguished:

the preprocessing or text normalisation;
the natural language processing, doing linguistic analysis and annotation;
the calculation of acoustic parameters, which translates the linguistically annotated symbolic structure into a table containing only physically relevant parameters;
and the synthesis, transforming the parameter table into an audio file.