Proceedings of the International Conference on Applied Science and Technology on Social Science 2021 (iCAST-SS 2021)

Developing Automatic English Speaking Skills Testing System Using Speech Recognition

Authors
Aliv Faizal M.*, aliv@pens.ac.id
Creative Multimedia Technology Department, Politeknik Elekronika Negeri Surabaya, Surabaya, Indonesia
Halimatus Sa’diyahhalimatus@pens.ac.id
Creative Multimedia Technology Department, Politeknik Elekronika Negeri Surabaya, Surabaya, Indonesia
Elizabeth Anggraeni Amalolisa@pens.ac.id
Department of Electronics Engineering, Politeknik Elekronika Negeri Surabaya, Surabaya, Indonesia
Salim Nabhansalimnabhan@unipa.ac.id
English Language Education Department, Universitas PGRI Adi Buana, Surabaya, Indonesia
M.H. Assidqihasbi@pens.ac.id
Creative Multimedia Technology Department, Politeknik Elekronika Negeri Surabaya, Surabaya, Indonesia
Imam Dui Agussalimimam_dui@pens.ac.id
Department of Electronics Engineering, Politeknik Elekronika Negeri Surabaya, Surabaya, Indonesia
Corresponding Author
Aliv Faizal M.aliv@pens.ac.id
Available Online 4 March 2022.
DOI
10.2991/assehr.k.220301.095How to use a DOI?
Keywords
computer assisted language testing; speech recognition; machine learning; job interview simulation; computer assisted language learning
Abstract

English teachers have been testing students speaking skill through student presentation and other kinds of direct testing. Testing the speaking ability of a big number of students is time consuming. Hence, an automatic model of testing English speaking skill is demanded. A computer assisted testing of English speaking using the technology of speech recognition comes handy. In this paper, we propose a tool to automatically score a student’s English-speaking performance. The proposed system applies the speech recognition technology with the string-matching algorithm in PHP language to process the voice input of the test candidate for scoring. The development resulted two main products namely the web-based speaking test app, and the data set for the scoring system. The initial stage focus of the development is the trained data model for the scoring system. The data was tested using Confusion Matrix, and it resulted percentage in accuracy of 80, precision of 84, recall and sensitivity of 94, and F1 Score of 88. This concludes that the app can help the researcher enrich the data and refine the score for a better automatic English-speaking testing system. Hence, once the app and the data model are perfected, it is ready for efficiency in English speaking testing.

Copyright
© 2022 The Authors. Published by Atlantis Press SARL.
Open Access
This is an open access article under the CC BY-NC license.

Download article (PDF)

Volume Title
Proceedings of the International Conference on Applied Science and Technology on Social Science 2021 (iCAST-SS 2021)
Series
Advances in Social Science, Education and Humanities Research
Publication Date
4 March 2022
ISBN
10.2991/assehr.k.220301.095
ISSN
2352-5398
DOI
10.2991/assehr.k.220301.095How to use a DOI?
Copyright
© 2022 The Authors. Published by Atlantis Press SARL.
Open Access
This is an open access article under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Aliv Faizal M.
AU  - Halimatus Sa’diyah
AU  - Elizabeth Anggraeni Amalo
AU  - Salim Nabhan
AU  - M.H. Assidqi
AU  - Imam Dui Agussalim
PY  - 2022
DA  - 2022/03/04
TI  - Developing Automatic English Speaking Skills Testing System Using Speech Recognition
BT  - Proceedings of the International Conference on Applied Science and Technology on Social Science 2021 (iCAST-SS 2021)
PB  - Atlantis Press
SP  - 577
EP  - 584
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.220301.095
DO  - 10.2991/assehr.k.220301.095
ID  - M.2022
ER  -