Abstract: Features are disclosed for automatically identifying a speaker. Artifacts of automatic speech recognition (“ASR”) and/or other automatically determined information may be processed against individual user profiles or models. Scores may be determined reflecting the likelihood that individual users made an utterance. The scores can be based on, e.g., individual components of Gaussian mixture models (“GMMs”) that score best for frames of audio data of an utterance. A user associated with the highest likelihood score for a particular utterance can be identified as the speaker of the utterance. Information regarding the identified user can be provided to components of a spoken language processing system, separate applications, etc.

Abstract: Methods and systems to build containers of uncategorized items are described. In some example embodiments, the methods and systems collect information from titles of items available for purchase within a network-based publication system, generate a title for a container of uncategorized items available for purchase within the network-based publication system that is based on the collected information, and associate two or more uncategorized items available for purchase within the network-based publication system to the container based on a comparison of titles of the two or more uncategorized items and the generated title for the container of uncategorized items.

Abstract: Features are disclosed for generating intent-specific results in an automatic speech recognition system. The results can be generated by utilizing a decoding graph containing tags that identify portions of the graph corresponding to a given intent. The tags can also identify high-information content slots and low-information carrier phrases for a given intent. The automatic speech recognition system may utilize these tags to provide a semantic representation based on a plurality of different tokens for the content slot portions and low information for the carrier portions. A user can be presented with a user interface containing top intent results with corresponding intent-specific top content slot values.

Abstract: Power consumption for a computing device may be managed by one or more keywords. For example, if an audio input obtained by the computing device includes a keyword, a network interface module and/or an application processing module of the computing device may be activated. The audio input may then be transmitted via the network interface module to a remote computing device, such as a speech recognition server. Alternately, the computing device may be provided with a speech recognition engine configured to process the audio input for on-device speech recognition.

Abstract: Features are disclosed for automatically identifying a speaker. Artifacts of automatic speech recognition (“ASR”) and/or other automatically determined information may be processed against individual user profiles or models. Scores may be determined reflecting the likelihood that individual users made an utterance. The scores can be based on, e.g., individual components of Gaussian mixture models (“GMMs”) that score best for frames of audio data of an utterance. A user associated with the highest likelihood score for a particular utterance can be identified as the speaker of the utterance. Information regarding the identified user can be provided to components of a spoken language processing system, separate applications, etc.

Abstract: Provided are systems and methods for receiving a plurality of item submissions from a plurality of mobile user devices (each item submission of the plurality of item submissions including: item identifier data indicative of an item; and item location data indicative of a location of the item), determining a determined location for the item (using the respective item location data for each of the plurality of item submissions), and storing the determined location for the item in an item location database. The determined location for the item is stored in association with an item identifier corresponding to the item, and the item location database stores determined locations for a plurality of items.

Abstract: Methods and systems to build containers of uncategorized items are described. In some example embodiments, the methods and systems collect information from titles of items available for purchase within a network-based publication system, generate a title for a container of uncategorized items available for purchase within the network-based publication system that is based on the collected information, and associate two or more uncategorized items available for purchase within the network-based publication system to the container based on a comparison of titles of the two or more uncategorized items and the generated title for the container of uncategorized items.

Abstract: Methods and systems to build containers of uncategorized items are described. In some example embodiments, the methods and systems collect information from titles of items available for purchase within a network-based publication system, generate a title for a container of uncategorized items available for purchase within the network-based publication system that is based on the collected information, and associate two or more uncategorized items available for purchase within the network-based publication system to the container based on a comparison of titles of the two or more uncategorized items and the generated title for the container of uncategorized items.

Abstract: Features are disclosed for generating intent-specific results in an automatic speech recognition system. The results can be generated by utilizing a decoding graph containing tags that identify portions of the graph corresponding to a given intent. The tags can also identify high-information content slots and low-information carrier phrases for a given intent. The automatic speech recognition system may utilize these tags to provide a semantic representation based on a plurality of different tokens for the content slot portions and low information for the carrier portions. A user can be presented with a user interface containing top intent results with corresponding intent-specific top content slot values.

Abstract: Features are disclosed for automatically identifying a speaker. Artifacts of automatic speech recognition (“ASR”) and/or other automatically determined information may be processed against individual user profiles or models. Scores may be determined reflecting the likelihood that individual users made an utterance. The scores can be based on, e.g., individual components of Gaussian mixture models (“GMMs”) that score best for frames of audio data of an utterance. A user associated with the highest likelihood score for a particular utterance can be identified as the speaker of the utterance. Information regarding the identified user can be provided to components of a spoken language processing system, separate applications, etc.

Abstract: Features are disclosed for generating intent-specific results in an automatic speech recognition system. The results can be generated by utilizing a decoding graph containing tags that identify portions of the graph corresponding to a given intent. The tags can also identify high-information content slots and low-information carrier phrases for a given intent. The automatic speech recognition system may utilize these tags to provide a semantic representation based on a plurality of different tokens for the content slot portions and low information for the carrier portions. A user can be presented with a user interface containing top intent results with corresponding intent-specific top content slot values.

Abstract: Systems and methods of providing composite search results are disclosed. A search query may be received by a general purpose search engine of a first website from a user. At least one general purpose search result may be determined based on the search query using the general purpose search engine. The at least one general purpose search result may be based on a search of multiple websites. Item listing information of an item listed for sale on a second website may be determined based on the search query. The at least one general purpose search result and the item listing information may be caused to be displayed concurrently on the first website to the user. The user may be enabled to interact with the second website regarding the item by interacting with the item listing information displayed on the first website.

Abstract: Features are disclosed for automatically identifying a speaker. Artifacts of automatic speech recognition (“ASR”) and/or other automatically determined information may be processed against individual user profiles or models. Scores may be determined reflecting the likelihood that individual users made an utterance. The scores can be based on, e.g., individual components of Gaussian mixture models (“GMMs”) that score best for frames of audio data of an utterance. A user associated with the highest likelihood score for a particular utterance can be identified as the speaker of the utterance. Information regarding the identified user can be provided to components of a spoken language processing system, separate applications, etc.

Abstract: Method and system is described to effectuate a single action upload of images from a mobile device. When the system detects activation of a visual control provided by an on-line service, the system activates a camera provided with the mobile device of the user, detects a resulting image captured by the camera, and uploads the resulting image from the mobile device of the user to a destination computing device. The destination computing device can be a server system associated with the on-line service.

Abstract: Power consumption for a computing device may be managed by one or more keywords. For example, if an audio input obtained by the computing device includes a keyword, a network interface module and/or an application processing module of the computing device may be activated. The audio input may then be transmitted via the network interface module to a remote computing device, such as a speech recognition server. Alternately, the computing device may be provided with a speech recognition engine configured to process the audio input for on-device speech recognition.

Abstract: Features are disclosed for automatically identifying a speaker. Artifacts of automatic speech recognition (“ASR”) and/or other automatically determined information may be processed against individual user profiles or models. Scores may be determined reflecting the likelihood that individual users made an utterance. The scores can be based on, e.g., individual components of Gaussian mixture models (“GMMs”) that score best for frames of audio data of an utterance. A user associated with the highest likelihood score for a particular utterance can be identified as the speaker of the utterance. Information regarding the identified user can be provided to components of a spoken language processing system, separate applications, etc.

Abstract: A system and method of metadata refinement using behavioral patterns is disclosed. In some embodiments, user behavioral data for results of a search query is received. The results can include an untagged item and a plurality of tagged items. A determination can then be made that the tagged items have been assigned a plurality of types of metadata. The untagged item can then be identified as a candidate to be tagged with at least one of the plurality of types of metadata assigned to the tagged items. In some embodiments, the user behavioral data comprises clickstream data indicating that a user selected the untagged item and at least one of the tagged items during a single search event.

Abstract: Features are disclosed for automatically identifying a speaker. Artifacts of automatic speech recognition (“ASR”) and/or other automatically determined information may be processed against individual user profiles or models. Scores may be determined reflecting the likelihood that individual users made an utterance. The scores can be based on, e.g., individual components of Gaussian mixture models (“GMMs”) that score best for frames of audio data of an utterance. A user associated with the highest likelihood score for a particular utterance can be identified as the speaker of the utterance. Information regarding the identified user can be provided to components of a spoken language processing system, separate applications, etc.

Abstract: Power consumption for a computing device may be managed by one or more keywords. For example, if an audio input obtained by the computing device includes a keyword, a network interface module and/or an application processing module of the computing device may be activated. The audio input may then be transmitted via the network interface module to a remote computing device, such as a speech recognition server. Alternately, the computing device may be provided with a speech recognition engine configured to process the audio input for on-device speech recognition.