Abstract: In non-limiting examples of the present disclosure, systems, methods and devices for resolving temporal ambiguities are presented. A natural language input may be received. A temporal component of the input may be identified. A determination may be made that the temporal component includes a conjunction that separates temporal meeting block alternatives. A temporal ambiguity may be identified in one of the meeting block alternatives. A plurality of syntax tree permutations may be generated for the meeting block alternative where the ambiguity was identified. A machine learning model that has been trained to identify a most relevant permutation for a given natural language input may be applied to each of the plurality of permutations. A temporal meeting block alternative corresponding to the most relevant permutation may be surfaced.

Abstract: Systems and methods for managing models for call data are disclosed. For example, the system may include one or more memory units storing instructions and one or more processors configured to execute the instructions to perform operations. The operations may include receiving, from a user device, an input indicating a segment of a first recorded call and an attribute associated with the segment. The operations may include determining a parameter of a model, wherein the model is associated with the attribute. The operations may include changing the parameter based on the input. The operations may include generating an updated model based on the changed parameter, wherein the updated model may be configured to analyze recorded calls having one or more segments with the same attribute.

Abstract: A technique making use of a few-shot model to determine whether a query text content belongs to a same language as a small set of examples, or alternatively provide a next member in the same language to the small set of examples. The related few-shot model makes use of convolutional models that are trained in a “learning-to-learn” fashion such that the models know how to evaluate few-shots that belong to the same language. The term “language” in this usage is broader than spoken languages (e.g., English, Spanish, German, etc.). “Language” refers to a category, or data domain, of expression through characters. Belonging to a given language is not specifically based on what the language is, but the customs or traits expressed in that language.

Abstract: A door cover defines an upper surface of a door of a refrigerator. The door cover includes a microphone pocket in which a microphone mount is provided and a separate speaker pocket in which a speaker mount is provided.

Abstract: To generate substantially condition-invariant and speaker-discriminative features, embodiments are associated with a feature extractor capable of extracting features from speech frames based on first parameters, a speaker classifier capable of identifying a speaker based on the features and on second parameters, and a condition classifier capable of identifying a noise condition based on the features and on third parameters. The first parameters of the feature extractor and the second parameters of the speaker classifier are trained to minimize a speaker classification loss, the first parameters of the feature extractor are further trained to maximize a condition classification loss, and the third parameters of the condition classifier are trained to minimize the condition classification loss.

Abstract: A computer-implemented technique is described herein for expediting a user's interaction with a digital assistant. In one implementation, the technique involves receiving a system prompt generated by a digital assistant in response to an input command provided by a user via an input device. The technique then generates a predicted response based on linguistic content of the system prompt, together with contextual features pertaining to a circumstance in which the system prompt was issued. The predicted response corresponds to a prediction of how the user will respond to the system prompt. The technique then selects one or more dialogue actions from a plurality of dialogue actions, based on a confidence value associated with the predicted response. The technique expedites the user's interaction with the digital assistant by reducing the number of system prompts that the user is asked to respond to.

Abstract: Techniques and architectures for implementing a team of virtual assistants are described herein. The team may include multiple virtual assistants that are configured with different characteristics, such as different functionality, base language models, levels of training, visual appearances, personalities, and so on. The characteristics of the virtual assistants may be configured by trainers, end-users, and/or a virtual assistant service. The virtual assistants may be presented to end-users in conversation user interfaces to perform different tasks for the users in a conversational manner. The different virtual assistants may adapt to different contexts. The virtual assistants may additionally, or alternatively, interact with each other to carry out tasks for the users, which may be illustrated in conversation user interfaces.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for machine translation using neural networks. In some implementations, a text in one language is translated into a second language using a neural network model. The model can include an encoder neural network comprising a plurality of bidirectional recurrent neural network layers. The encoding vectors are processed using a multi-headed attention module configured to generate multiple attention context vectors for each encoding vector. A decoder neural network generates a sequence of decoder output vectors using the attention context vectors. The decoder output vectors can represent distributions over various language elements of the second language, allowing a translation of the text into the second language to be determined based on the sequence of decoder output vectors.

Abstract: An electronic furniture assembly of the present invention comprises: (i) a furniture assembly comprising: (A) a base (e.g., a seat portion), (B) at least one transverse member (e.g., a side, armrest or backrest), and (C) a coupler for selectively coupling the base to the transverse member; (ii) an electrical hub for providing electrical power to the speakers, the hub configured to selectively reside within the furniture assembly; and (iii) a speaker system mounted within one or more portions of the furniture assembly. The speaker system comprises one or more speakers mounted within the base and transverse member, hiding the speakers therein, and saving space within a home or office, using the same footprint for both furniture and speakers, providing a high fidelity surround sound system.

Abstract: Embodiments described herein provide for a voice biometrics system execute machine-learning architectures capable of passive, active, continuous, or static operations, or a combination thereof. Systems passively and/or continuously, in some cases in addition to actively and/or statically, enrolling speakers. The system may dynamically generate and update profiles corresponding to end-users who contact a call center. The system may determine a level of enrollment for the enrollee profiles that limits the types of functions that the user may access. The system may update the profiles as new contact events are received or based on certain temporal triggering conditions.

Abstract: Systems and methods for providing a translation-enabled multiparty communication session utilize a translation service that is configured to receive individual audio streams from each participant in the communication session, and which then generates a participant text stream for each participant based on their audio stream. This can be accomplished using speech-to-text techniques. Each participant's text stream is then translated into a different language to create a translated participant text stream. Individual participants in the communication session can then subscribe to the translated participant text streams to view a translation of what each participant is saying during the communication session.

Abstract: A method and system for operating a multitenancy telephony system including receiving a plurality of call requests associated with an application server; selecting a subset of the received call requests using a load balancer; assigning each selected call request to one of a plurality of resources creating, on the assigned resource, a call session for each selected call request; and coordinating the created call sessions resulting from the selected subset of the received call requests associated with the application server.

Abstract: An example method comprising receiving a plurality of categories of availability statuses, receiving a plurality of categories of different contact groups, identifying, for each contact, at least one contact group, receiving a plurality of rules for call handling, each of the plurality of rules being associated with at least one of the different availability statuses and at least one of the different contact groups, receiving a telephone call from a first caller, identifying a caller identifier of the first caller, determining a contact group of the first caller, determining a current availability status of the recipient, identifying an applicable rule based on the contact group of the first caller and the current availability status of the recipient, selecting an audio prompt based on the applicable rule, and presenting the audio prompt to the first caller based on the applicable rule, the audio prompt indicating a preference of communication.

Abstract: There is a growing problem in correctional facility telecommunications systems in which parties on a voice call may connect inmate callers with restricted parties. Prison communication systems monitor calls to prevent such activity, but in Voice over Internet Protocol (VoIP) environments such systems may fail to detect this activity. The present disclosure provides details of a system and method for using SIP messages common in VoIP environments to detect illicit activity initiated by a party on a voice call within a controlled environment. Scenarios are detected in which a called party connects an inmate caller to a restricted party via three-way call conferencing, call forwarding, or other call features. Corrective actions are then taken when such activity is detected, such as call blocking or alerting officials illicit activity is occurring.

Abstract: A translation system receives examples of translations between a first language and a second language. In response to receiving request to translate a source text from the first language to the second language, the system ranks the examples based on the example's applicability to one or more portions of the source text. The system performs additional training of a neural network that was pre-trained to translate from the first language to the second language, where the additional training is based on one or more top-ranking examples. The system translates the source text to the second language using the additionally trained neural network.

Abstract: The present invention relates to a method for performing a detection of a topic of a message introduced in a real-time customer service messaging platform. In the method a message comprising at least one word from which the topic is definable is received; a topic from the received message is extracted; it is inquired from a database if the topic is determinable from a number of messages received chronically earlier than the received message; and an indication is generated to an operator of the real-time customer service messaging platform in accordance with a detection result obtained through an inquiry to the database. Some aspects of the present invention relate to a network node, to a computer program product and to a system.

Abstract: In an approach to generating conversational image representations, one or more computer processors detect one or more utterances by a user, wherein utterances are either textual or acoustic. The one or more computer processors generate one or more image representations of the one or more detected utterances utilizing a generative adversarial network restricted by one or more user privacy parameters, wherein the generative adversarial network is fed with an extracted sentiment, a generated avatar, an identified topic, an extracted location, and one or more user preferences. The one or more computer processors display the generated one or more image representations on one or more devices associated with respective one or more recipients of the one or more utterances.

Abstract: Computer-implemented methods, computer program products, and computer systems for detecting and preventing exploitative calls. The computer-implemented methods include one or more processors configured for receiving voice call data corresponding to an active voice call with a user device, wherein the voice call data comprises caller voice data and user voice data; determining that the caller voice data includes a solicitation of information from a user associated with the user device; retrieving user profile data comprising user health data associated with the user; determining user call stress levels based on the voice call data and the user profile data; and responsive to determining that the user call stress levels exceed a predetermined threshold and that the solicitation of information includes a material information request, generating an alert.

Abstract: A system and method for customizing communication processing within a communication platform that includes configuring a service handler, which comprises of at least setting execution code of the service handler; setting an association between the service handler and a communication event of the communication platform; detecting the occurrence of the communication event of a communication facilitated through the communication platform; invoking the service handler in response to the event, which comprises: executing the execution code in a serverless execution environment and generating a service handler response; and augmenting the communication in accordance with the service handler response.

Abstract: The present disclosure is directed to a method and system for routing communications. A plurality of communications-enabled devices are coupled to a communications network. Each communications-enabled device has a device identifier. A user-programmable database associates each of a plurality of user identifiers with a device identifier. A router circuit is coupled to the communications network through a portal communication device having a portal identifier. A user designates a destination user identifier when making a communication. The router circuit switches the call, directing the call to a destination device identifier responsive to the user-programmable database association between the destination user identifier and a device identifier. According to certain aspects, communications are directed to users and are routed to associated equipment, a user identifier serving as a virtual extension number.