In this study, the UMU platform is used to provide teaching resources for English speaking courses, and the iSmart platform is used for speaking practice and assessment to realize the problem of accurate teaching of spoken English in elementary school. A genetic algorithm is introduced to establish a learning model for content organization, and personalized organization and recommendation of learning content is achieved by calculating the initial population and fitness value, and setting termination conditions. An output-oriented teaching system centered on “teaching concept-teaching assumption-teaching process” is established to strengthen the synergy between language input and output, and to improve students’ oral expression ability, grammatical accuracy and fluency. The findings show that the experimental group’s oral performance is significantly higher than that of the control group (MD=4.12), and in terms of accuracy and fluency, the experimental group (78.32±8.45) is higher than that of the control group (73.47±7.61).82.2% of the surveyed students indicated that the innovative teaching mode made their independent learning easier, and the experimental group students’ performance in contextualization analysis is higher (P<0.05). The results confirm the effectiveness of the proposed innovative model in stimulating learning interest and enhancing comprehensive language ability in the new era educational environment, and provide an actionable practical path for elementary school English teaching reform.
