Using Speech Recognition Algorithms to Improve Listening Training Effects in Foreign

Abstract

Background: Understanding is a crucial element for accurate communication during foreign language learning. The learning results of conventional teaching strategies are often poor due to their very low preparation and response. There are more practicalities and advantages to speech recognition technology, which provides immediate responses, but its incorporation requires cautious study to safeguard its practicality and advantages. Method: In this paper, we gather speech signal datasets from 120 learners and divide them equally into two groups such as baseline and study group. In baseline group we provide a traditional technique but in study group, we proposed a new technique named artificial rabbit optimized hidden markov model (AR-HMM). Additionally, we made an analysis with proposed technique to improve the listening training effect in foreign English language. Result: As a result, first we used nine questionnaires to examine the participant's experience. We evaluate the performance of the proposed technique and compare it with an existing technique based on the parameters such as accuracy (95%), involvement (low (15%), medium (59%) and high (93%)), efficiency (90%) and time (14hr). Conclusion: When compared to the existing technique, our proposed technique has superior performance than others to improve listening training effect in foreign English language.