Speech transaction for blinds using speech-text-speech conversions

Abstract

Effective human computer interaction requires speech recognition and voice response.In this paper we present a concatenative Speech-Text- Speech(STS) system and discuss the issues relevant to the development of perfect human-computer interaction.The new STS system allows the visually impaired people to interact with the computer by giving and getting voice commands.Audio samples are collected from the individuals and then transcribed to text.A text file is used ,where the meanings for the transcribed texts are stored.In the synthesis phase,the sentences taken from the text file are converted to speech using unit selection synthesis.The proposed method leads to a perfect human-computer interaction. © Springer-Verlag Berlin Heidelberg 2011.