Digital transformation in education demands an interactive, flexible learning system focused on developing competencies in the era of Society 5.0. However, preliminary research indicates that current teaching materials do not yet accommodate students' needs in learning to speak. The speech text material contained in the teaching materials has not integrated the Deep Learning approach and the use of technology is still very limited, resulting in minimal participation and boring learning. This results in the material being presented textually and underutilizing visual aspects, resulting in a lack of in-depth understanding of the material and students' speaking competence not being achieved. This study aims to test the validity of teaching materials for persuasive speech text speaking skills based on the Deep Learning approach assisted by Mobile Learning. This study uses the Sugiyono development model with seven research stages. The validation process was carried out by media experts, material experts, and teachers using a validation sheet as an instrument. The results showed that the media expert's validation obtained a score of 89.58%, the material expert obtained a score of 100%, and the teacher obtained a score of 97.92%, so that the overall average score was 95.83% with a very valid category. Thus, this teaching material is declared very valid for use in teaching speaking skills on persuasive speech text material in class VIII of junior high school and can be continued to the implementation stage to test its effectiveness and impact on student learning outcomes.
Copyrights © 2026