Abstract: Techniques for improved image classification are provided. Face embeddings are generated for each face depicted in a collection of images, and the face embeddings are clustered based on the individual whose face is depicted. Based on these clusters, each embedding is assigned a label reflecting the cluster assignments. Some or all of the face embeddings are then used to train a classifier model to generate cluster labels for new input images. This classifier model can then be used to process new images in an efficient manner, and classify them into appropriate clusters.

Abstract: A system configured to vary a speech speed of speech represented in input audio data without changing a pitch of the speech. The system may vary the speech speed based on a number of different inputs, including non-audio data, data associated with a command, or data associated with the voice message itself. The non-audio data may correspond to information about an account, device or user, such as user preferences, calendar entries, location information, etc. The system may analyze audio data associated with the command to determine command speech speed, identity of person listening, etc. The system may analyze the input audio data to determine a message speech speed, background noise level, identity of the person speaking, etc. Using all of these inputs, the system may dynamically determine a target speech speed and may generate output audio data having the target speech speed.

Abstract: Categorizing images may include training a first neural network to cluster a plurality of images to obtain a first image embedding space, wherein a vector representation is determined for each of the plurality of images based on the training, determining a vector norm value corresponding to each of the plurality of images based on the vector representation for each of the plurality of images, and identifying a first subset of the images for which a corresponding vector norm value satisfies a predetermined vector norm quality threshold. Then, a second neural network may be trained using the first subset of images to obtain a second image embedding space, and the second image embedding space may be used to categorize additional images.

Abstract: A single communication application can display an initial modality view associated with the group communication on a canvas of a communication application user interface. The initial modality view can be one of multiple available views. The single communication application can receive a selection to display a different from the initial modality view. Each active user in the group communication can be on a different computing device with different active instances of the single communication application which are each configured to switch independently of each other active instances of the single communication application running on different computing devices to modality views that are different from the initial modality view. The communication application can display the selected modality view on the canvas by transitioning from the initial modality view.

Abstract: A system configured to vary a speech speed of speech represented in input audio data without changing a pitch of the speech. The system may vary the speech speed based on a number of different inputs, including non-audio data, data associated with a command, or data associated with the voice message itself. The non-audio data may correspond to information about an account, device or user, such as user preferences, calendar entries, location information, etc. The system may analyze audio data associated with the command to determine command speech speed, identity of person listening, etc. The system may analyze the input audio data to determine a message speech speed, background noise level, identity of the person speaking, etc. Using all of these inputs, the system may dynamically determine a target speech speed and may generate output audio data having the target speech speed.

Abstract: A system configured to vary a speech speed of speech represented in input audio data without changing a pitch of the speech. The system may vary the speech speed based on a number of different inputs, including non-audio data, data associated with a command, or data associated with the voice message itself. The non-audio data may correspond to information about an account, device or user, such as user preferences, calendar entries, location information, etc. The system may analyze audio data associated with the command to determine command speech speed, identity of person listening, etc. The system may analyze the input audio data to determine a message speech speed, background noise level, identity of the person speaking, etc. Using all of these inputs, the system may dynamically determine a target speech speed and may generate output audio data having the target speech speed.

Abstract: A single communication application can display an initial modality view associated with the group communication on a canvas of a communication application user interface. The initial modality view can be one of multiple available views. The single communication application can receive a selection to display a different from the initial modality view. Each active user in the group communication can be on a different computing device with different active instances of the single communication application which are each configured to switch independently of each other active instances of the single communication application running on different computing devices to modality views that are different from the initial modality view. The communication application can display the selected modality view on the canvas by transitioning from the initial modality view.

Abstract: A communication application displays a modality view that may be one of a collaboration, a gallery, or a messaging view on a conversation canvas. The application may display an initial view according a modality selection by the user. A user is enabled to select a next view from the set by providing a user action such as a tap, swipe action, etc. The application dynamically generates the next modality view according to the initial view. Common participants and common contexts are used to configure the next view. Subsequent to configuration, the application displays the next modality view on the conversation canvas by transitioning from the initial view. The application retains session information from the initial view to restore the initial view session if the user selects to return to the initial view.

Abstract: A communication application displays a modality view that may be one of a collaboration, a gallery, or a messaging view on a conversation canvas. The application may display an initial view according a modality selection by the user. A user is enabled to select a next view from the set by providing a user action such as a tap, swipe action, etc. The application dynamically generates the next modality view according to the initial view. Common participants and common contexts are used to configure the next view. Subsequent to configuration, the application displays the next modality view on the conversation canvas by transitioning from the initial view. The application retains session information from the initial view to restore the initial view session if the user selects to return to the initial view.

Abstract: A communication application displays a modality view that may be one of a collaboration, a gallery, or a messaging view on a conversation canvas. The application may display an initial view according a modality selection by the user. A user is enabled to select a next view from the set by providing a user action such as a tap, swipe action, etc. The application dynamically generates the next modality view according to the initial view. Common participants and common contexts are used to configure the next view. Subsequent to configuration, the application displays the next modality view on the conversation canvas by transitioning from the initial view. The application retains session information from the initial view to restore the initial view session if the user selects to return to the initial view.

Abstract: A communication application displays a modality view that may be one of a collaboration, a gallery, or a messaging view on a conversation canvas. The application may display an initial view according a modality selection by the user. A user is enabled to select a next view from the set by providing a user action such as a tap, swipe action, etc. The application dynamically generates the next modality view according to the initial view. Common participants and common contexts are used to configure the next view. Subsequent to configuration, the application displays the next modality view on the conversation canvas by transitioning from the initial view. The application retains session information from the initial view to restore the initial view session if the user selects to return to the initial view.