Bi-Lingual Text to Speech Synthesis System for Urdu and Sindhi

Text to Speech Synthesis along with the Speech Recognition is widely used throughout the world to enhance the accessibility of the information and enable even the disabled persons to interact with the computers in order to get the potential benefit from this high-tech revolution. In this paper we introduce a bi-lingual novel algorithm for the synthesis of Urdu and Sindhi language text. The devised bi-lingual algorithm uses knowledge based approach along with the hybrid rule based and concatenative acoustic methods to provide efficient and accurate conversion of Urdu and Sindhi text into the high quality speech. The algorithm has been implemented in the VB programming language with a GUI based interface. The proposed system works with high accuracy and has a great potential to be used for variety of applications. The system is versatile enough and can be used for speech recognition also.