Myanmar Text-to-Speech Processing

With fast development of information technology in Myanmar, people need to access with text-to-speech system of Myanmar language.

This research presents a framework for building Myanmar Language text-to-speech system, all the essential issues in developing the text-to speech for Myanmar text analysis, prosody generation and speech synthesis.

Text-to-speech synthesis is a module or system or machine that converts the input text into the acoustic speech signal that people can understand.

The goal is to produce a text-to-speech synthesis of Myanmar language that can synthesize a natural sound.

Natural Language Processing (NLP) is gradually becoming a more multidisciplinary field, and research in remotely connected aspects of language such as TTS may benefit from as well as contribute to some area of (NLP).

  Myanmar synthesized speeches of HMM and DNN based systems

  Myanmar synthesized speeches of DNN, LSTM, and Hybrid-LSTM based systems

  Myanmar synthesized speeches of DNN and Hybrid-LSTM based systems with different input features