Abstract: Among other things, embodiments of the present disclosure may be used to help train speech recognizers for improving generalized voice experience quality in a chat bot system. In some embodiments, the system provides users with games to play to increase user engagement with the chat bot system.

Abstract: Systems and methods are disclosed for response retrieval using communication session vectors. In one implementation, a first communication session is received. The first communication session includes a first communication. The first communication session is encoded as a first vector. A second vector is identified within a defined proximity of the first vector. The second vector represents a second communication session that includes a second communication. The second communication is provided within the first communication session in response to the first communication.

Abstract: Incompatible item pairings may be eliminated or at least reduced when multiple items are presented. A pairwise approach is taken to train a machine learning model to return an incompatibility score for any given pair of items, which indicates a degree of incompatibility between the pair of items. Once trained, the machine learning model may be used to determine an incompatibility score for each unique pairing of items in a set of multiple items. In some embodiments, a graph is generated having nodes that correspond to the multiple items and undirected edges between pairs of the nodes. Scores are generated for each edge of the graph, a minimum spanning tree in the graph is determined, and the items are ranked based at least in part on the minimum spanning tree so that the items can be presented according to the ranking.

Abstract: Systems and methods may be used to process and output information related to a non-speech vocalization, for example from a user attempting to mimic a non-speech sound. A method may include determine a mimic quality value associated with an audio file by comparing a non-speech vocalization to a prerecorded audio file. For example, the method may include determining an edit distance between the non-speech vocalization and the prerecorded audio file. The method may include assigning a mimic quality value to the audio file based on the edit distance. The method may include outputting the mimic quality value.

Abstract: Methods, systems, and computer programs are presented for providing a user interface (UI) for monitoring and debugging an Artificial Intelligence (AI) chatting hot. One method includes operations for receiving a selection on the UI to replay an electronic conversation between a first and a second party, selecting conversation data associated with the electronic conversation from a data log having conversation data from several electronic conversations, and analyzing the conversation data to identify conversation parameters. The conversation parameters include text in each entry of the electronic conversation, timing of the entries, and debugging parameters for each entry. The method further includes an operation for causing presentation of the electronic conversation on the UI, which includes presenting the text of each entry, the audio corresponding to speech associated with each entry timed according to the timing of the entry, and the debugging parameters embedded within the presented text.

Abstract: Examples of the present disclosure describe systems and methods relating to conversational system user behavior identification. A user of the conversational system may be evaluated based on one or more factors. The one or more factors may be compared to an aggregated measure for a larger group of conversational system users, such that “anomalous” behavior (e.g., behavior that deviates from a normal behavior) may be identified. When a user is identified as exhibiting anomalous behavior, the conversational system may adapt its interactions with the user in order to encourage, discourage, or further observe the identified behavior. As a result, the conversational system may be able to verify a user's anomalous behavior, discourage the anomalous behavior, or take other action while interacting with the user.

Abstract: Systems and methods are disclosed for response retrieval using communication session vectors. In one implementation, a first communication session is received. The first communication session includes a first communication. The first communication session is encoded as a first vector. A second vector is identified within a defined proximity of the first vector. The second vector represents a second communication session that includes a second communication. The second communication is provided within the first communication session in response to the first communication.

Abstract: Systems and methods are disclosed for conversationally-relevant content requests. In one implementation, a first communication is processed to identify a first content element within the communication. The first content element is stored within a first content repository. A content request is generated based on the first content repository. The content request is provided to the user. A second communication is received from the user in response to the content request. An action is initiated based on the second communication.

Abstract: Examples of the present disclosure describe systems and methods relating to conversational system user experience. In an example, a conversational system may use one or more sensors of a user device to affect the topic or direction of a conversation or to identify a new conversation topic. The conversational system may also receive input from a user, wherein a GUI may enable the user to specify or alter semantic information used during the conversation. The GUI may comprise one or more skeuomorphic elements designed to provide a familiar or intuitive way for the user to interact with the conversational system. The GUI may also be used to disambiguate messages or convey emotion or sentiment to the user. In another example, haptic or audio feedback may be provided alongside a message to convey emotion to the user during the conversation.

Abstract: Examples of the present disclosure describe systems and methods relating to conversational system user behavior identification. A user of the conversational system may be evaluated based on one or more factors. The one or more factors may be compared to an aggregated measure for a larger group of conversational system users, such that “anomalous” behavior (e.g., behavior that deviates from a normal behavior) may be identified. When a user is identified as exhibiting anomalous behavior, the conversational system may adapt its interactions with the user in order to encourage, discourage, or further observe the identified behavior. As a result, the conversational system may be able to verify a user's anomalous behavior, discourage the anomalous behavior, or take other action while interacting with the user.

Abstract: Examples of the present disclosure describe systems and methods relating to conversational system user behavior identification. A user of the conversational system may be evaluated based on one or more factors. The one or more factors may be compared to an aggregated measure for a larger group of conversational system users, such that “anomalous” behavior (e.g., behavior that deviates from a normal behavior) may be identified. When a user is identified as exhibiting anomalous behavior, the conversational system may adapt its interactions with the user in order to encourage, discourage, or further observe the identified behavior. As a result, the conversational system may be able to verify a user's anomalous behavior, discourage the anomalous behavior, or take other action while interacting with the user.

Abstract: Representative embodiments disclose mechanisms for discovering, synthesizing, presenting and interacting with multiple chat bots. A broker system can interact with a user and receive a query which is passed to a search engine. The search results are evaluated for bots provided by domains. The query can be submitted to the discovered bots. For domains that do not provide a bot, a bot can be synthesized if desired and the query submitted to the synthesized bots. Answers from the bots are retrieved. Answer are directly presented for bots that are directly surfaced to the user in the search results page. Answers can be fused for bots that are not directly surfaced to the user and presented via a broker bot. Answers from the one bot can be fed to other bots so that the bots can participate in a group chat between bots and the user.

Abstract: Among other things, embodiments of the present disclosure may be used to help train speech recognizers for improving generalized voice experience quality in a chat bot system. In some embodiments, the system provides users with games to play to increase user engagement with the chat bot system.

Abstract: Systems and methods are provided for augmenting or annotating image data to help a user locate an item among a number of items in a physical environment. The systems disclosed herein can access an image of a location that includes a number of items. The systems may then identify the items and the location of the items within the image. Further, the systems may create an interactive alternative visualization of the identified items to facilitate a user locating a particular item. For example, the alternative visualization may include presenting an ordered set of thumbnails corresponding to the items. The user can select the desired item using the alternative visualization and the systems can annotate a location within the image where the item can be located, thereby enabling the user to locate the item in the physical environment.

Abstract: Systems and methods for automated mathematical chatting. The systems and methods convert any identified non-numerical inputs into vectors and then perform the mathematical equation utilizing the vectors instead of the nonnumeric inputs along with any other identified numeric inputs to obtain a numerical vector result. The systems and methods decode the numerical vector result into a result feature and then search one or more databases for output based on the result feature. The systems and methods provide the selected output from the one or more databases in response to the mathematical query.

Abstract: Techniques for detecting dimensions of an object from a three dimensional (3D) image may be provided. Dimensions of an object may be determined based upon a received 3D image and calibration information for calibrating a dimension related to the 3D image. The calibration information may be utilized for calibrating aspect ratios of the 3D image. In an example, a recommended item may be identified and presented to the user based on the determined dimensions of the object.

Abstract: Systems and methods are provided for augmenting or annotating image data to help a user locate an item among a number of items in a physical environment. The systems disclosed herein can access an image of a location that includes a number of items. The systems may then identify the items and the location of the items within the image. Further, the systems may create an interactive alternative visualization of the identified items to facilitate a user locating a particular item. For example, the alternative visualization may include presenting an ordered set of thumbnails corresponding to the items. The user can select the desired item using the alternative visualization and the systems can annotate a location within the image where the item can be located, thereby enabling the user to locate the item in the physical environment.

Abstract: Systems and methods are provided for determining dimensional criteria of a physical space and identifying a set of items that satisfy the dimensional criteria. For example, the dimensional criteria may be determined based on an image depicting the physical space or search terms provided by a user. Composite keys may be generated based on the dimensional criteria and one or more search terms provided by the user, and the generated composite keys may be used to search one or more databases storing sorted item keys and corresponding item identifiers.

Abstract: A method, apparatus, and/or system for providing an action with respect to a mobile device using a robotic device that tracks the user. In accordance with at least one embodiment, a request to perform an action with respect to an electronic device is received. Information may be sent to one or more robotic devices within a proximity of the electronic device. A robotic device of the one or more robotic devices may be selected to perform the action. An indication may be received from the robotic device that indicates that the user has interacted with the robotic device. Instructions may be sent to the robotic device to perform the action with respect to the electronic device. A location of the user may be tracked while charging is performed by the robotic device. The robotic device may be instructed to follow the user at a threshold distance from the user.

Abstract: Systems and methods are provided for augmenting or annotating image data to help a user locate an item among a number of items in a physical environment. The systems disclosed herein can access an image of a location that includes a number of items. The systems may then identify the items and the location of the items within the image. Further, the systems may create an interactive alternative visualization of the identified items to facilitate a user locating a particular item. For example, the alternative visualization may include presenting an ordered set of thumbnails corresponding to the items. The user can select the desired item using the alternative visualization and the systems can annotate a location within the image where the item can be located, thereby enabling the user to locate the item in the physical environment.