Metadata Project – TVB Experience
5
7. Subtitle Preparation Terminals Most of the content waiting digitisation have subtitle files associated. The subtitle file contains VITC time information of the each subtitle lines, which is required for automatic subtitle play out. However, due to the reality that VITC would becomes unavailable after digitisation, and need to use frame count as timing information instead. The Subtitle Preparation system was developed to convert VITC timing information to FC (Frame Count) timing information. Since frame count starts from 0, the value of the first frame count of the converted FC information will be set to zero and the rest of the FC will be aligned automatically.
Metadata Input System (MIS)
The MIS was developed using Microsoft Visual Studio 2005 Professional, and database servers are running on SQL Server 2008. Two database servers are running in mirror mode to achieve one to one back up configuration. There are more than 90 workstations for metadata input. They are running Windows XP Chinese to support not only Traditional Chinese but also Japanese, Korean and Simplified Chinese metadata input. Metadata of Episode, Scene and Item levels are basically in Traditional Chinese, however, Simplified Chinese, Korean and Japanese are used for metadata occasionally.
User requirements were collected from different departments in March 2010, to identify the metadata needed for their daily operation. Software coding commenced in April 2010 and beta version became available in June 2010 for fine tuning. The beta version was released for soft launch in August 2010. The software was official launched in February 2011.
Metadata Input
Currently, there are more than 30 staff working on metadata input. It is observed that around ten hours is required to complete one episode metadata input which has 81 data fields. The MIS has been enhanced to assist metadata input in order to increase input efficiency, such as pull down menus which contain generic data applicable to the entire programme and capturing of data from other systems etc.
Future Development
Currently more than 25,000 programme hours of digitised content have metadata created, a Basic Metadata Search Engine (BMSE) has been developed to assist Research, Programme and Production colleagues to find their desired materials from the digitised content archive. Proxy of the found content will be available for preview purposes, LTO tape number of the archive (non-compressed) is also available when a user decides to use it for broadcast or programming. The BMSE will be released to certain users for trial next month. Thumbnail generation for the found content is a feature to be included in the next development stage.
Research on ways to improve metadata input efficiency continues. It includes the study of using voice recognition methods to input Chinese metadata and interface with other in-house metadata sources to reduce the necessity of re-entering duplicate data.
Snap Shots of the Four MIS input main pages
Diagram 1: Programme level metadata input main page
Diagram 2: Episode level metadata input main page
Diagram 3: Item level metadata input main page
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48