Speech Synthesizer for English Audio with Indian Accent

Abstract

The main aim of the project is to generate synthetic Speech in Indian accent. The synthesizer will take input as a transcript and produce audio file as a output. Primarily we will learn theoretical aspects of speech synthesis and apply them to develop complete code to synthesis audio. We will use Text processing which is responsible for determining all knowledge about the text that is not specifically phonetic and Phonetic analysis which focuses on the phone level within each word, tagging each phone with information about what sound to produce and how to produce it. General issues such as NLP, DSP, different voices, accents and multiple languages, HMM model, phonemes are discussed. As we know Indian Accent is different from that of American or British Accent. Hence our focus is on Indian accent. This would largely help even for those people who cannot understand the American accent. The speech synthesizer has enormous applications such as reading for blind people, telecommunication services, language education, aid to handicapped person, talking books and toys, call centre automation.