摘要:
A computer-implemented method of indexing a speech lattice for search of audio corresponding to the speech lattice is provided. The method includes identifying at least two speech recognition hypotheses for a word which have time ranges satisfying a criteria. The method further includes merging the at least two speech recognition hypotheses to generate a merged speech recognition hypothesis for the word.
摘要:
Systems and methods for packaging, delivery, and use of digital content. In one embodiment, a license matrix is created and used for evaluating the license rights applicable to a requested use of a content item. The license matrix can comprise a plurality of licenses grouped in license chains, and each license can include a set of entitlement rules and a set of grants. A license chain can be associated with an identifier corresponding to a use of content. In another embodiment, it is provided a method of packaging unsecured content for secured delivery over a computer network. In yet another embodiment, a system and method are provided for applying digital management rules to a content item at the time a user requests a use of the content item. In some embodiments, systems and methods allow evaluation of a requested use against a group of licenses to determine the applicable license rights for the requested use.
摘要:
Systems and methods are provided for archiving and querying data updates received or generated by an update processing system associated with an electronic catalog. In some embodiments, item data updates associated with items represented in an electronic catalog are formed into new objects consisting of collections of records of item data updates. The new objects may have keys and versions associated with each record within them. The use of these objects allows item data updates, which are generally on the order of a few kilobytes, to be efficiently stored in archive storage systems, and to be queried without affecting the performance of the system that hosts the electronic catalog.
摘要:
Disclosed are various embodiments for using contributor-provided attributes in conjunction with an electronic item catalog. A computer reconciles a plurality of attributes for an item in an electronic catalog using feedback data related to the item. Each of the attributes is provided by one of a plurality of merchants. The reconciled attribute of the item is updated in the electronic catalog.
摘要:
Various embodiments of a system and method for conditioning search results are described. The system and method for conditioning search results may include a search result conditioning component. The search result conditioning component may be configured to determine a search result group including multiple search results each corresponding to an item offered for sale; such search results may be results of a search query. The search result conditioning component may be configure to modify the search result group by removing one or more search results from the search result group in response to determining that a measure of similarity between the one or more search results and other search results of the search result group is greater than a requisite measure of similarity (e.g., a threshold). The search result conditioning component may be configured to output the modified search result group.
摘要:
A system and method are provided for determining document relevance determination to a selected category for a document contained within a linked network of documents. The network may be represented by a network map including nodes representing documents and edges representing links between the documents. The method may include identifying each node in the network map known to belong to the selected category, identifying each node known to be outside of the selected category, and identifying nodes having an unknown category. The method may also include assigning a category rank based on the node category identification, identifying each link from each node and each link to each node, and assigning link weights based on the identified links. The method may additionally include determining node relevance to the selected category based on the assigned category rank and the assigned link weights. An origination domain and a destination domain for each link may be determined such that link weights are assigned for both incoming and outgoing links for each node.