Automated speech recognition – captions on recorded teaching materials


January 27th, 2021

The automatic speech recognition (ASR) feature of the Engage (Echo360) platform has been switched on as default. This means transcripts are automatically provided for newly recorded or uploaded videos.

From September 2021 the provision of a transcript or closed captions will be a requirement for all lecture material that is pre-recorded or recorded live, and further details will be provided to staff in due course.

In the interim, the following guidance about ASR has been prepared to support staff who use the Engage platform, and is also relevant to other transcript services we use (e.g. that within Microsoft Teams).

  1. There is no expectation that staff will routinely spend time editing transcripts, nor is doing so currently considered a viable option in the future. We will continue to lobby the software provider to improve the quality of their ASR feature.
  2. Staff should be aware that the quality of the ASR transcription, in common with other similar services at the University and elsewhere, can vary. Students have been notified of this, and informed that they should not consider any transcript on its own to be a reliable source of academic content.
  3. Where a member of staff has concerns about the quality of an ASR transcript of their recording, the transcript may be deleted. Details of how to do this can be found in the Engage Staff Moodle page.

For further help and support, staff can contact the Engage team: engage@nottingham.ac.uk.

Leave a Reply

Other

Need news? See you on SharePoint

After 14 years of service, Campus News is being retired as the university’s staff news platform.  […]

Roads and car parks closed for refurbishing work

As part of ongoing road improvements at the university, works will be taking place to resurface […]