A Preliminary Work on Speech to Speech Translation
Priyanka Padmane1, Ayush Pakhale2, Sagar Agrel3, Ankita Patel4, Sarvesh Pimparkar5, Prajwal Bagde6
[1] Professor, Dept. of Computer Technology, Priyadarshini College of Engineering, Nagpur - India
[2][3][4][5[6] Student Dept. of Computer Technology, Priyadarshini College of Engineering, Nagpur - India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The automatic translation with one human language into another is referred to as "machine translation." The main purpose is to bridge the linguistic gap between people who speak different languages, communities, or countries. There are 18 main language and ten scripts that are frequently used. Because the majority of Indians, particularly remote peasants, cannot understand, read, or write English, an excellent language translator is required. Machine translation systems that transform source text to another will assist Indians in living in a more enlightened society without language barriers. Because English is a worldwide language and Hindi is the language spoken by the majority of Indians, we propose an English to Hindi machine translation system based on recurrent neural networks (RNN), LSTM (Long short-term memory), and attention processes. The automatic translation of one natural language into another is referred to as "machine translation." The main purpose is to bridge the linguistic gap between people who speak different languages, communities, or countries. There are 18 official languages and ten scripts that are frequently used. The majority of Indians, particularly isolated peasants, do not understand, read, or write English, necessitating the implementation of an effective language translator. Machine translation systems that convert text from one language to another will help Indians live in a more enlightened society that is free of language barriers.
Key Words: RNN, LSTM, Speech to text, text to Speech, Multi linguistic.
