Speech To Speech (S2s) User Interface Design For Emerging Artificial Intelligence Markets In Education

Abstract

Socio-economic development has represented a variety of challenges, such as providing access to information, bridging communication gaps and technological and functional illiteracy. Significant advances have been made in the field of spoken language technology development for under –resourced languages, such as text-to-speech (TTS) and Speech to text (STT) systems. Speech recognition is expanding capabilities to support application in emerging markets.

More people around the world will gain the ability to use speech enable application in day-to-day activity. This experiment designs and develops speech to speech (S2S) interface for the calculator appliance. The MFCC is used for the feature extraction mode for the system. The system is tested based on a neural network approach.

The possibility for the designing of a software system of speech-to-speech recognition using one of the techniques of artificial intelligence applications neuron networks where this system is used for real time education-based speech UI design.  The UI is created utilizing speech as input information and result turn into a speech. The people groups in developing business sector require a speech empower machine for them without hands free communication. The system is tested over the speaker independent mode. Performance of the system achieved 96.75% with minimum real time factor.