Transcribear – Introducing An Online Automated Transcription & Annotation Tool

Yuhua Chen

12 Feb 2020 16:00-17:00, GE403

Reliable transcription and annotation (also known as “coding”) is essential for research in many areas of Humanities and Social Sciences, where data is often collected through interviews, focus groups, meeting or classroom observations, or many other types of recordings. Such recordings would then need to be transcribed into text to make the data readable and searchable, and yet transcription is notoriously time-consuming and challenging. A good tool is therefore important in facilitating this type of speech to text conversion.

Transcribear” was developed to cater for the needs of an easy-to-use tool which integrates automated/manual transcription, audio/video playback, spelling checks, and various shortcut keys into one online interface. To minimize human errors, the functionality of annotation validation in the format of tags <> is also added. A technical paper which documented earlier development of this online tool has been published in the Journal of Digital Scholarship in the Humanities (DSH). Originally designed for a multimodal corpus project (CAWSE), this browser-based application can be customized for individual users’ needs in terms of the annotation scheme and corresponding shortcut keys.

In this talk, the co-founder of Transcribear, Yu-Hua will 1) introduce the development of speech to text technology and its applications in audio transcription; 2) demonstrate how audio transcription tools can make our research work easier while also improve the quality of outputs; 3) discuss various issues from her experience of transcribing and annotating large amounts of written and spoken data.

Slides for talk can be downloaded below: