A lot of interesting things come from IBM Research. Something new that I found particularly interesting is the IBM Multimodal Annotation Tool. The tool assists in annotating MPEG files with MPEG-7 meta data (both video and audio. Each shot in the video sequence can be annotated with static scene descriptions, key object descriptions, event descriptions, and other lexicon sets. Audio segments can be delimited and described as well. The annotated descriptions are stored as MPEG-7 descriptions in an XML file. The tool can also open MPEG-7 files in order to display the annotations for the corresponding MPEG file.

• The W3C Multimodal Architecture, Part 1: Overview and challenges
• The W3C Multimodal Architecture, Part 2: The XML specification stack
• The W3C Multimodal Architecture, Part 3: A multimodal Web service