A system to support accurate transcription of information systems lectures for disabled students

Despite the substantial progress that has been made in the area of Automatic Speech Recognition, the performance of current systems is still below the level required for accurate transcription of lectures. This paper explores a different approach focusing on automation of the editing process of lecture transcripts produced by ASR software. The resultant Semantic and Syntactic Transcription Analysing Tool, based on natural language processing and human interface design techniques, is a step forward in the production of meaningful postlecture materials, with minimal investment in time and effort by academic staff and responds to the challenge of meeting the needs of students with disabilities. This paper reports on the results of a study to assess the potential of SSTAT to make the transcription process of Information Systems lectures more efficient and to determine the level of correction required to render the transcripts usable by students with a range of disabilities.