Abstract: A method and apparatus for tagging image content with rich metadata is provided. The metadata is generated from keyword descriptions of image content spoken by human taggers while viewing the content. Voice recognition software is employed to identify the key keywords in an audio stream and the resultant metadata is associated in a synchronous manner with the relevant image content. A control console allows the human tagger to rapidly navigate onscreen menus and select different taglines for providing multilevel metadata tagging of the image content. An integrated system provides for the storage of tagged digital image content, with near immediate access to tagged raw footage for viewing and editing, and for easy searching and accessing of finalized footage. A method of serving the tagged content is also provide, which allows the content to be streamed over the web at an acceptable image resolution while maintaining the associated metatags.

Abstract: A method and apparatus for tagging image content with rich metadata is provided. The metadata is generated from keyword descriptions of image content spoken by human taggers whilst viewing the content. Voice recognition software is employed to identify the key keywords in an audio stream and the resultant metadata is associated in a synchronous manner with the relevant image content. A control console allows the human tagger to rapidly navigate onscreen menus and select different taglines for providing multilevel metadata tagging of the image content. An integrated system provides for the storage of tagged digital image content, with near immediate access to tagged raw footage for viewing and editing, and for easy searching and accessing of finalized footage. A method of serving the tagged content is also provide, which allows the content to be streamed over the web at an acceptable image resolution whilst maintaining the associated metatags.