An update on automatic transcription vs. Manual transcription

Frank E. Ritter, Catherine Bouyat, Kaitlyn Ekdahl, Daniel Guzek

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Cognitive modelers have long used verbal protocol analysis to gather data to test their models. Recently developed tools offer support for automatic transcription of audio. In this short paper, we compare the time it takes to transcribe a video done (a) exclusively with Google’s subtitle tool, (b) corrected from Google’s subtitle tool, and (c) done completely by hand. We found that using the subtitle tool alone can yield too high an error rate, correcting Google subtitles took about 2.5x the video length, and transcribing completely by hand took approximately 11x the video length. We can thus recommend using Google subtitles as a starting point for verbal transcription as it offers a useful speed up in transcription. In addition, we can recommend when and how to use Google subtitles, and the use of headphones and automatic spelling correction in text editors.

Original languageEnglish (US)
Title of host publicationProceedings of ICCM 2016 - 14th International Conference on Cognitive Modeling
EditorsDavid Reitter, Frank E. Ritter
PublisherThe Pennsylvania State University
Pages268-269
Number of pages2
ISBN (Electronic)9780998508207
StatePublished - 2016
Event14th International Conference on Cognitive Modeling, ICCM 2016 - University Park, United States
Duration: Aug 3 2016Aug 6 2016

Publication series

NameProceedings of ICCM 2016 - 14th International Conference on Cognitive Modeling

Conference

Conference14th International Conference on Cognitive Modeling, ICCM 2016
Country/TerritoryUnited States
CityUniversity Park
Period8/3/168/6/16

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Modeling and Simulation

Fingerprint

Dive into the research topics of 'An update on automatic transcription vs. Manual transcription'. Together they form a unique fingerprint.

Cite this