Ulster University Logo

Broadcast Language Identification System (BLIS)

Connolly, James, Curran, K, McKevitt, P, Macrae, J and Craig, S (2014) Broadcast Language Identification System (BLIS). In: Proc. of the 16th Irish Machine Vision and Image Processing Conference (IMVIP-14), University of Ulster, Magee, Derry/Londonderry, Northern Ireland. University of Ulster. 2 pp. [Conference contribution]

[img] PDF - Published Version

URL: http://hdl.handle.net/2262/71411


Virtually all TV transmission systems use highly automated file-based broadcast systems for audio and video. Content management systems automatically deliver programmes ready for broadcast by matching content to a schedule and deliver all ancillary services in their correct format and on time. Within these sub-systems is a growing need to validate that the correct language is delivered to a particular service and/or region. In many cases, a single instance of a programme exists and the automated system merely selects the correct language for a particular service. This process is currently managed manually by operators listening to the audio of each programme and confirming that the accompanying language is correct for its video broadcast. Incorrect language transmission can be caused by system faults or errors in the scheduling workflow. An error can occur at numerous points during the broadcast. The Broadcast Language Identification System (BLIS) will provide a single operator with the ability to monitor multiple services by “dash-boarding” language flags from each service and enable the operator to intervene if an error is detected. BLIS will examine streaming audio from a pre-broadcast to identify spoken language within the broadcast content and compare it with the expected language of the video for broadcast.

Item Type:Conference contribution (Poster)
Faculties and Schools:Faculty of Computing & Engineering
Faculty of Computing & Engineering > School of Computing and Intelligent Systems
Faculty of Arts
Faculty of Arts > School of Creative Arts and Technologies
Research Institutes and Groups:Computer Science Research Institute > Intelligent Systems Research Centre
Computer Science Research Institute
ID Code:30742
Deposited By: Professor Paul McKevitt
Deposited On:08 Jan 2015 15:13
Last Modified:08 Jan 2015 15:13

Repository Staff Only: item control page