Application of RNN-ET-Based Digital Education Using Speech Interactions

In order to address this evolution, six tendencies are identified in this system. The tendencies are still there and will continue to have an impact on educational technology. The physicality of interactions and the learner's physical environment first emerged as genuine elements of digital education. The line separating the physical and the digital has vanished. Similar to how integrated learning scenarios have absorbed the conflict between individual and social views on cognition, AIED now places more emphasis on social interactions than it did in the beginning. Another development is the processing of learners' behavioral particles, which individually do not have a lot of semantic meaning but are predictive of knowledge states when processed by machine learning techniques on big data sets. Another trend has emerged as a result of the development of probabilistic models and the incorporation of crowdsourcing techniques: learning environment design is now less deterministic than it once was. The idea of a learning environment has developed from a somewhat tight box to an open ecosystem where various stakeholders can connect and where various components are dispersed across various platforms. Teachers now interact not only during the design phase (authoring), but also during the runtime phase (orchestration), which is significant to note among these stakeholders. Preprocessing, feature extraction, and model performance evaluation are all used in the suggested method. Normalization is used for preprocessing. It uses the Mel Filter Bank and DCT for feature extraction. It employs the RNN-ET Model for gauging the model's efficacy. The proposed method is evaluated in comparison to the RNN and the CNN models. When compared to more simplistic approaches, it excels.