Abstract: Embodiments are configured to provide one or more candidates based in part on an input identifier associated with an input interface. In an embodiment, a system includes an input method editor (IME) that can be configured to provide one or more candidates based in part on a prediction model and one or more input scope values associated with an input interface. The IME can also operate to store an entered string including one or more associated input scope values to a storage component. The stored information can be used when suggesting candidates.

Abstract: A system and method of refining context-free grammars (CFGs). The method includes deriving back-off grammar (BOG) rules from an initially developed CFG and utilizing the initial CFG and the derived BOG rules to recognize user utterances. Based on a response of the initial CFG and the derived BOG rules to the user utterances, at least a portion of the derived BOG rules are utilized to modify the initial CFG and thereby produce a refined CFG. The above method can carried out iterativey, with each new iteration utilizing a refined CFG from preceding iterations.

Abstract: Architecture for integrating and generating back-off grammars (BOG) in a speech recognition application for recognizing out-of-grammar (OOG) utterances and updating the context-free grammars (CFG) with the results. A parsing component identifies keywords and/or slots from user utterances and a grammar generation component adds filler tags before and/or after the keywords and slots to create new grammar rules. The BOG can be generated from these new grammar rules and can be used to process the OOG user utterances. By processing the OOG user utterances through the BOG, the architecture can recognize and perform the intended task on behalf of the user.

Abstract: Architecture for integrating and generating back-off grammars (BOG) in a speech recognition application for recognizing out-of-grammar (OOG) utterances and updating the context-free grammars (CFG) with the results. A parsing component identifies keywords and/or slots from user utterances and a grammar generation component adds filler tags before and/or after the keywords and slots to create new grammar rules. The BOG can be generated from these new grammar rules and can be used to process the OOG user utterances. By processing the OOG user utterances through the BOG, the architecture can recognize and perform the intended task on behalf of the user.

Abstract: Asynchronous voice messaging is employed to enable individuals to send voice clips back and forth over a network. In at least some embodiments, the voice clips reside in the form of discrete, manageable objects that can be easily managed by the user. In addition, some embodiments described provide hand-free/eyes-free functionality that can permit a user to create and send (or receive and play) a voice message without having to physically enter an intended recipient's address or otherwise digitally manipulate the device over which the message is sent or received.

Abstract: The presently described subject matter allows the user to interactively browse a collection of electronic files, such as a digital photo collection, remotely using voice activated mobile device. Files from the collection of electronic files can be displayed on the mobile device or they can be directed from the mobile device to be sent to a remote display device, where the file can be displayed. Also, a user can record voice and text annotations and keywords from the mobile device to preserve with an electronic file, and which is transported along with the file when it is displayed.