Abstract: One or more computing devices, systems, and/or methods for auto-generating subject suggestions are provided. For example, a set of n-grams are generated from text within an email body of an email being composed by a user and/or from descriptive information of attachments to the email. Features of the set of n-grams are identified, such as a date feature, a proper noun feature, an activity feature, a number of words feature, etc. Classifiers are used to assign probabilities to the n-grams based upon the features. A probability may specify a strength of an n-gram in relation to a user preference of the user for an email subject derived from the n-gram. In this way, an email subject suggestion is generated based upon a target n-gram with a probability exceeding a threshold. The email subject suggestion is provided to the user, such as during email composition or during sending of the email.

Abstract: One or more computing devices, systems, and/or methods for auto-generating subject suggestions are provided. For example, a set of n-grams are generated from text within an email body of an email being composed by a user and/or from descriptive information of attachments to the email. Features of the set of n-grams are identified, such as a date feature, a proper noun feature, an activity feature, a number of words feature, etc. Classifiers are used to assign probabilities to the n-grams based upon the features. A probability may specify a strength of an n-gram in relation to a user preference of the user for an email subject derived from the n-gram. In this way, an email subject suggestion is generated based upon a target n-gram with a probability exceeding a threshold. The email subject suggestion is provided to the user, such as during email composition or during sending of the email.

Abstract: One or more computing devices, systems, and/or methods for auto-generating subject suggestions are provided. For example, a set of n-grams are generated from text within an email body of an email being composed by a user and/or from descriptive information of attachments to the email. Features of the set of n-grams are identified, such as a date feature, a proper noun feature, an activity feature, a number of words feature, etc. Classifiers are used to assign probabilities to the n-grams based upon the features. A probability may specify a strength of an n-gram in relation to a user preference of the user for an email subject derived from the n-gram. In this way, an email subject suggestion is generated based upon a target n-gram with a probability exceeding a threshold. The email subject suggestion is provided to the user, such as during email composition or during sending of the email.

Abstract: A user device is enabled by an audio-visual assistant for audio-visual interaction with a user. The audio-visual assistant enables the user device to track the user's eyes and face to determine objects on the screen that the user is currently observing. Various tasks can be executed on the objects based on further input provided by the user. The user can provide further inputs via facial gestures, voice or combinations thereof for executing the various tasks.

Abstract: A user device is enabled by an audio-visual assistant for audio-visual interaction with a user. The audio-visual assistant enables the user device to track the user's eyes and face to determine objects on the screen that the user is currently observing. Various tasks can be executed on the objects based on further input provided by the user. The user can provide further inputs via facial gestures, voice or combinations thereof for executing the various tasks.

Abstract: A user device is enabled by an audio-visual assistant for audio-visual interaction with a user. The audio-visual assistant enables the user device to track the user's eyes and face to determine objects on the screen that the user is currently observing. Various tasks can be executed on the objects based on further input provided by the user. The user can provide further inputs via facial gestures, voice or combinations thereof for executing the various tasks.

Abstract: One or more computing devices, systems, and/or methods for auto-generating subject suggestions are provided. For example, a set of n-grams are generated from text within an email body of an email being composed by a user and/or from descriptive information of attachments to the email. Features of the set of n-grams are identified, such as a date feature, a proper noun feature, an activity feature, a number of words feature, etc. Classifiers are used to assign probabilities to the n-grams based upon the features. A probability may specify a strength of an n-gram in relation to a user preference of the user for an email subject derived from the n-gram. In this way, an email subject suggestion is generated based upon a target n-gram with a probability exceeding a threshold. The email subject suggestion is provided to the user, such as during email composition or during sending of the email.

Abstract: A user device is enabled by an audio-visual assistant for audio-visual interaction with a user. The audio-visual assistant enables the user device to track the user's eyes and face to determine objects on the screen that the user is currently observing. Various tasks can be executed on the objects based on further input provided by the user. The user can provide further inputs via facial gestures, voice or combinations thereof for executing the various tasks.