THE EFFECTIVENESS OF THE ELSA SPEAK APPLICATION IN TEACHING PRONUNCIATION AMONG STUDENTS AT MA. AN-NIDHOMIYAH

Irda Andini (English Education Department, STKIP PGRI Bangkalan)
Siti Maria Ulfa (English Education Department, STKIP PGRI Bangkalan)

SECONDARY : Jurnal Inovasi Pendidikan Menengah (JIPM)
Vol. 6 No. 3 (2026)

ABSTRACT This study aimed to examine the effectiveness of the ELSA Speak application in improving the English pronunciation skills of students at MA An-Nidhomiyah. The study was motivated by the persistent challenges in students’ pronunciation accuracy, which are influenced by first language (L1) interference, limited opportunities for individualized feedback, and low self-confidence during oral communication when conventional teaching methods are employed. A quantitative approach was adopted using a quasi-experimental design with a One-Group Pre-test/Post-test Design. The population consisted of all students at MA An-Nidhomiyah during the 2025–2026 academic year, while the sample comprised 27 eleventh-grade students selected through purposive sampling. Data were collected using pre-test and post-test pronunciation assessments and analyzed with SPSS 25 through validity, reliability, normality, and paired-samples t-test analyses. The findings revealed a statistically significant improvement in students’ pronunciation performance following the implementation of the ELSA Speak application, as evidenced by significant differences between the pre-test and post-test scores. This study provides empirical evidence regarding the effectiveness of artificial intelligence-assisted pronunciation instruction within the context of an Islamic senior secondary school (Madrasah Aliyah), a setting that remains underrepresented in previous studies. The findings suggest that ELSA Speak has the potential to serve as an effective instructional medium by providing objective real-time feedback, enhancing pronunciation accuracy, reducing foreign language anxiety, and promoting greater learner autonomy in English language learning. Furthermore, the results offer practical implications for integrating AI-assisted learning into English pronunciation instruction at the secondary education level. ABSTRAK Penelitian ini bertujuan untuk menganalisis efektivitas penggunaan aplikasi ELSA Speak dalam meningkatkan kemampuan pronunciation bahasa Inggris siswa di MA An-Nidhomiyah. Penelitian ini dilatarbelakangi oleh masih rendahnya akurasi pengucapan siswa yang dipengaruhi oleh interferensi bahasa pertama (first language/L1), keterbatasan pemberian umpan balik secara individual, serta rendahnya kepercayaan diri siswa ketika melakukan komunikasi lisan menggunakan metode pembelajaran konvensional. Penelitian menggunakan pendekatan kuantitatif dengan desain quasi-experimental tipe One-Group Pre-test/Post-test Design. Populasi penelitian adalah seluruh siswa MA An-Nidhomiyah tahun ajaran 2025–2026, sedangkan sampel penelitian berjumlah 27 siswa kelas XI yang dipilih menggunakan teknik purposive sampling. Data dikumpulkan melalui tes pre-test dan post-test kemampuan pronunciation, kemudian dianalisis menggunakan perangkat lunak SPSS 25 melalui uji validitas, reliabilitas, normalitas, dan paired-samples t-test. Hasil penelitian menunjukkan adanya peningkatan kemampuan pronunciation yang signifikan setelah penerapan aplikasi ELSA Speak, yang dibuktikan oleh perbedaan skor sebelum dan sesudah perlakuan. Temuan ini memperluas implementasi pembelajaran berbasis artificial intelligence pada konteks Madrasah Aliyah, yang masih relatif terbatas dalam penelitian sebelumnya. Dengan demikian, aplikasi ELSA Speak dapat menjadi alternatif media pembelajaran yang efektif untuk meningkatkan akurasi pengucapan, menyediakan umpan balik yang objektif dan bersifat real-time, mengurangi kecemasan berbahasa, serta mendorong kemandirian belajar siswa dalam pembelajaran bahasa Inggris.