World Wide Telecom Web Search

摘要

Searching of spoken content has been of interest to the speech community for varied reasons in the past. The call centers are interested in searching information the calls where an agent may not have provided appropriate answer to the customer. Video search is mostly dependent on the accompanying audio analysis. In such scenarios, the query interface is still a visual interaction and multiple results can be presented to the user, with effective browsing and presentation controls. We challenge ourselves to enable searching of user generated audio content through a audio-only query-result interface. The motivation to perform such a search, the challenges in terms of data, interface and users are presented in this talk. The hope is that the audience will be able to identify sub-problems in this large space of WWTW search.