Speech to speech interaction system using Multimedia Tools and Partially Observable Markov Decision Process for visually impaired students

In general, visually impaired students need of another person's to teach them with the help of computers and book. However, a number of students are not aware of using the computers and understanding the concepts by self. In order to solve this issue, a speech to speech interaction system is developed on the basis of a novel dialogue management system. This interaction is developed by combining Multimedia tools and Partially Observable Markov Decision Process (POMDP) with agenda based model used in the proposed dialogue management system to learn the speech signals from user and system will reply accordingly. The proposed system helps visually impaired students to learn easily using a novel dialogue management system. Word Error Rate, Recognition cum retrieval rate and Misrecognition Retrieval Rate are calculated for the proposed POMDP with Agenda Based dialogue management system. The experimental results are compared with Finite-State Based dialogue management system, Frame Based dialogue management system, and Probabilistic dialogue management system. The experimental results proved that the good performance of the proposed POMDP with Agenda Based dialogue management system. The proposed model is trained with 125 speakers out of which 46 were visually impaired and tested with 95 untrained speakers out of which 32 are visually impaired.