Presentation

The Effectiveness of Computer Programs in the Transcription and Analysis of Spoken Discourse: Towards a Protocol for Pronunciation Corpora

---

Abstract

The analysis of spoken language has always had to address two main issues: the tedious method of extracting variables and researchers’ potential subjectivity when analyzing the data. The use of digital tools/software programs afford partial automatization of processes and the potential for increased objectivity. While these tools have improved analysis processes, some programs are more suitable for certain aspects of analysis than others. This paper provides an overview of five digital tools that have been recently employed in the analysis of spoken data: CLAN, ELAN, MFA, Phon, and Praat. The aim is to offer a brief but practical introduction to the usefulness of each program in relation to: transcribing audio files and extracting segmental (e.g., vowel length), prosodic (e.g., pitch range), and fluency (e.g., pause length) features. We hope that our protocols for transcription and analysis can also be used to further the ongoing goal of developing pronunciation corpora for use by the PSLLT community. The paper provides links to open source materials and information about each program through our website. We invite others to contribute to the website with their own handouts and resources.

How to Cite:

Ghanem, R., Edalatishams, I., Huensch, A., Puga, K. & Staples, S., (2019) “The Effectiveness of Computer Programs in the Transcription and Analysis of Spoken Discourse: Towards a Protocol for Pronunciation Corpora”, Pronunciation in Second Language Learning and Teaching Proceedings 11(1).