Literaturnachweis - Detailanzeige
| Autor/in | Daniels, Paul |
|---|---|
| Titel | Auto-Scoring of Student Speech: Proprietary vs. Open-Source Solutions |
| Quelle | In: TESL-EJ, 26 (2022) 3, (14 Seiten)Infoseite zur Zeitschrift
PDF als Volltext kostenfreie Datei |
| Zusatzinformation | ORCID (Daniels, Paul) |
| Sprache | englisch |
| Dokumenttyp | gedruckt; online; Zeitschriftenaufsatz |
| Schlagwörter | Speech Communication; Auditory Perception; Computer Uses in Education; Computer Assisted Testing; Scoring; Automation; Grading; Computer Mediated Communication; Feedback (Response); Learning Management Systems; Open Source Technology; Intelligibility; English (Second Language); Language Tests; Second Language Learning; Test of English as a Foreign Language |
| Abstract | This paper compares the speaking scores generated by two online systems that are designed to automatically grade student speech and provide personalized speaking feedback in an EFL context. The first system, "Speech Assessment for Moodle" ("SAM"), is an open-source solution developed by the author that makes use of Google's speech recognition engine to transcribe speech into text which is then automatically scored using a phoneme-based algorithm. "SAM" is designed as a custom quiz type for "Moodle," a widely adopted open-source course management system. The second auto-scoring system, "EnglishCentral," is a popular proprietary language learning solution which utilizes a trained intelligibility model to automatically score speech. Results of this study indicated a positive correlation between the speaking scores generated by both systems, meaning students who scored higher on the "SAM" speaking tasks also tended to score higher on the "EnglishCentral" speaking tasks and vice versa. In addition to comparing the scores generated from these two systems against each other, students' computer-scored speaking scores were compared to human-generated scores from small-group face-to-face speaking tasks. The results indicated that students who received higher scores with the online computer-graded speaking tasks tended to score higher on the human-graded small-group speaking tasks and vice versa. (As Provided). |
| Anmerkungen | TESL-EJ. e-mail: editor@tesl-ej.org; Web site: http://tesl-ej.org |
| Erfasst von | ERIC (Education Resources Information Center), Washington, DC |
| Update | 2024/1/01 |
Literaturbeschaffung und Bestandsnachweise in Bibliotheken prüfen
Permalink als QR-Code
Inhalt auf sozialen Plattformen teilen (nur vorhanden, wenn Javascript eingeschaltet ist)
Teile diese Seite:
