Posted by: Tom Nolle
content delivery networks, Google, speech recognition, video content, YouTube
Google has launched an automated captioning system for some of its partner channels, based on speech recognition algorithms. The move is aimed at making content more searchable and supporting access for the hearing-impaired, but it also could be the start of a system to textually index a lot of content to create the metadata need to categorize and cross-link complex video libraries.
Enterprises have wanted video indexing for some time and MPEG4 provides a metadata standard to store the information but there’s still the question of how it could be collected. Google may now have the answer, and if so there could be a new enterprise service coming out of the technology since the enterprise users list captioning and cross-linking as a major requirement for their own content and private content delivery networks (CDNs).