AUTOMATIC CAPTION FEATURES ON GOOGLE MEET AS A PRONUNCIATION ASSESSMENT TOOL
ABSTRACTThis study aims to find out the use Google Meet Automatic Caption feature to assist teachers of non-native English to assess their students' English pronunciation. We used a mixed-method with the explanatory-sequential approach following it. This research study was done at Tidar Univers...
Saved in:
Published in: | IJEE (Indonesian Journal of English Education) Vol. 9; no. 2; pp. 396 - 410 |
---|---|
Main Authors: | , , |
Format: | Journal Article |
Language: | English |
Published: |
27-12-2022
|
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | ABSTRACTThis study aims to find out the use Google Meet Automatic Caption feature to assist teachers of non-native English to assess their students' English pronunciation. We used a mixed-method with the explanatory-sequential approach following it. This research study was done at Tidar University with 12 participants, further reduced to 4 in the participant selection step. As the data were both quantitative and qualitative, we used the Word Error Rate (WER) formula quantitatively and Qualitative Content Analysis qualitatively. The findings show that Artificial Intelligence (AI) has a very sensitive system in transforming sounds into written forms. It also has an auto-correction system that sometimes can substitute a word with a meaningless one if a speaker pronounces the word unclearly or to the nearest word if a speaker mispronounces it. Even though it does not accurately process the punctuation and there is no sufficient correction on grammar, we believe the AI can help teachers in a pronunciation assessment.ABSTRAKPenelitian ini bertujuan untuk mengetahui penggunaan fitur Teks Otomatis pada Google Meet untuk membantu para guru Bahasa Inggris bukan penutur asli saat menilai pengucapan bahasa Inggris siswa mereka. Kami menggunakan metode penelitian campuran dengan pendekatan eksplanatori-sekuensial. Penelitian ini dilakukan di Universitas Tidar dengan jumlah peserta sebanyak 12 orang yang dikurangi menjadi 4 orang pada tahap seleksi peserta. Data penelitian bersifat kuantitatif dan kualitatif. Untuk menganalisa data secara kuantitatif digunakan rumus Word Error Rate (WER). Sedangkan, secara kualitatif mengunakan Analisis Isi Kualitatif. Hasil penelitian menunjukkan bahwa AI (Artificial Intelligence/ Kecerdasan buatan) memiliki sistem yang sangat sensitif dalam mengubah suara menjadi bentuk tulisan. Fitur ini juga memiliki sistem koreksi otomatis yang terkadang dapat menggantikan kata yang tidak berarti jika pembicara mengucapkan kata dengan tidak jelas atau diubah ke kata terdekat jika pembicara salah mengucapkan kata tersebut. Meskipun tidak memproses tanda baca secara akurat dan tidak ada koreksi yang memadai pada tata bahasa, kami yakin AI dapat membantu para guru dalam penilaian pengucapan. How to Cite: Ramadhika, B., Yosintha, R., Yunianti, S. S. (2022). Automatic Caption Features on Google Meet as a Pronunciation Assessment Tool. IJEE (Indonesian Journal of English Education), 9(2), 396-410. doi:10.15408/ijee.v9i2.22482 |
---|---|
ISSN: | 2356-1777 2443-0390 |
DOI: | 10.15408/ijee.v9i2.22482 |