Abstract: Systems and methods are disclosed for communication classification and escalation using a machine learning model. Metadata for a communication is obtained along with features of the content in the communication. Using a machine learning model, a likelihood that the communication comprises suspicious content or a topic of interest is determined from the metadata and or a communication transcript. If the likelihood that the communication comprises suspicious content or a topic of interest is at or above a predetermined threshold, then the communication is marked for review by a human agent; otherwise, the communication is ignored. A communication may have suspicious content or a topic of interest based upon the presence of keywords in the communication transcript either alone or in combined with metadata or other features of the content in the communication. The one or more keywords may be associated with illegal or unauthorized items, activities, or behaviors.

Abstract: In some implementations, one or more computers receive audio data indicating an utterance of a first participant during a call with a second participant. The one or more computers obtain text of the utterance of the first participant. The one or more computers identify an information card associated with the utterance based on the text of the utterance of the first participant. The one or more computers provide information card data causing the identified information card to be provided, for presentation during the call, by a device associated with the second participant.

Abstract: A method for routing customer service requests to call centers includes collecting data associated with customer service experience between a customer and a call center regarding a completed customer call. The collected data is analyzed to determine a quality of customer service for one or more completed calls between the customer and the call center. A nature of an incoming customer call is determined. The incoming customer call is routed to a call center based upon making reference to the analyzed collected data such that the nature of the customer call matches with corresponding one or more favorable attributes of the call center.

Abstract: Electronic communication methods and systems for determining sentiment associated with a communication, conveniently displaying indicia of the sentiment information, and scoring the sentiment information are disclosed. The methods and systems can include associating an emoji with a sentiment and annotating information, for example, a highlight reel or a waveform, with the emoji.

Abstract: A controllable grounded response generation framework includes a machine learning model, a grounding interface, and a control interface. The machine learning model is trained to output computer-generated text based on input text. The grounding interface is useable by the machine learning model to access a grounding source including information related to the input text. The control interface is useable by the machine learning model to recognize a control signal. The machine learning model is configured to include information from the grounding source in the computer-generated text and focus the computer-generated text based on the control signal.

Abstract: The present invention relates to communications methods and apparatus dynamically detecting and/or mitigating anomalies in communications systems/networks. An exemplary method embodiment includes the steps of: (i) storing a set of N key performance indicator (KPI) models; (ii) associating each of a plurality of recurring time slots of a recurring time frame on a per group member basis with one of the N KPI models wherein the associating including associating a first group member of a first group and a first recurring time slot with a first one of the N models, the first one of the N models being a first model; (iii) receiving event data for the first group member for a first time period; and (iv) determining based on the first model if a key performance indicator value for the first group member and first time period is anomalous.

Abstract: Contact centers strive to provide a positive and productive customer-agent interaction to successfully resolve the issue for a call. While audio content, such as music or messages, on hold are commonplace, selecting audio enhancements to be inserted into, and concurrently with, the customer-agent provides the customer and/or agent with cues and motivations to promote the successful completion of the call. Cues may be provided to announce the arrival or departure of an agent, virtually take a customer from one location to another for a different portion of the interaction, add excitement and anticipation to an upcoming event by providing an audio experience foreshadowing of the actual event, calm frayed nerves, or other purpose.

Abstract: One or more aspects of the disclosure relate to various functions and processes related to routing of calls based on origination and/or destination information. In one aspect, a method may comprise receiving a request for a call session, and the request may include an indicator that may indicate whether to route a call based on origination and/or destination information of the call session. Based on the indicator, a routing element may determine whether the call session should be routed based on origination information and/or destination information.

Abstract: Systems, apparatuses, methods, and computer program products are disclosed for predicting an entity, intent, and emotion based on weighted and segmented portions of captured speech. An example method includes capturing speech and separating the speech and/or text derived from the speech into a plurality of discrete segments. The example method further includes determining or generating metadata, the metadata an entity, an intent, an emotion, and a weight associated with each of the entity, the intent, the emotion. The example method further includes determining a next action based on the determined or generated entity, intent, and emotion of each of the plurality of discrete segments.

Abstract: Messaging with a virtual assistant by: receiving a first request (e.g., spoken by a user to a virtual assistant device hosting the virtual assistant) to check for messages received by an electronic device and transmitting the first request to a backend server for processing. The backend server requests data associated with the messages from the electronic device and, upon receiving the data, generates a response to the first request based on the received data. The backend server then transmits the response to the virtual assistant and the virtual assistant communicates the response (e.g., using audio produced by the virtual assistant device). The virtual assistant receives a reply from the user to the message. The reply is transmitted to the backend server for processing. The backend server transmits the reply to the electronic device which automatically transmits the reply to a recipient of the reply (e.g., a sender of the message).

Abstract: A method and apparatus for providing assistance to customers seeking assistance from agents of an enterprise is disclosed. The method includes receiving a signal indicating the caller's wish to speak with a live agent associated with the enterprise to seek the assistance from the live agent. In response to a receipt of the signal, it is determined whether at least one live agent from among a plurality of live agents associated with the enterprise is available for a voice interaction with the caller. Subsequent to determining that no live agent from among the plurality of live agents is currently available for the voice interaction with the caller, the caller is diverted to an asynchronous messaging channel. A textual messaging-based interaction is facilitated between a messaging agent and the caller on the asynchronous messaging channel to provide the assistance to the caller.

Abstract: A computer-implemented method, system, and computer program product for identifying requirements in a document. A document including requirements is received. Attribute related information in the document is identified using an attribute model. Component information in the document is identified using a component model. The attribute related information and the component information identified in the document are merged. Requirements in the document are identified from the merged attribute related information and component information. The requirements identified in the document are used to develop a product.

Abstract: The present application includes a method and system for multi-channel interaction. A communication session is initiated between a customer service representative (CSR) and an end user. Multi-channel communication is used between the end user and the CSR. The multi-channel communication includes at least voice and data. Information is presented to the end user via a user interface, and the user can confirm the accuracy of the information using the user interface.

Abstract: A caption of a multimodal message (e.g., social media post) can be identified as a named entity using an entity recognition system. The entity recognition system can use a visual attention based mechanism to generate a visual context representation from an image and caption. The system can use the visual context representation to identify one or more terms of the caption as a named entity.

Abstract: A media providing server and a method and computer program for providing call waiting content with respect to ARS order calls for a broadcast product provide call waiting content with respect to ARS order calls for the broadcast product and provide order content to enable a product to be ordered according to one of order methods except for an ARS order method.

Abstract: A system herein provides automated call-back of customers who have terminated an inbound call by exercising a call-back option of an interactive voice response unit or by abandoning the inbound call, using predictive modeling of caller value to prioritize call-backs. The call management system monitors the inbound customer call and detects any termination of the customer call. A call-back module opens a call-back record for the terminated customer call and associates that call-back record with an identified customer. The call-back module retrieves customer demographic data and other data associated with the identified customer. A predictive module determines a value prediction signal for the identified customer by modeling purchase and lapse behaviors and classifies each identified customer for either priority call-back or subordinate call-back treatment.

Abstract: A system, method, and/or device may receive a plurality of terms in a natural language of a user. The plurality of terms may relate to at least one domain of an enterprise of domains. The plurality of terms may be stored as text. For a (e.g., each) term of the plurality of terms, one or more domain levels may be determined in which the term is associated. The domain levels (e.g., each of the domain levels) may be associated with a specificity. From the domain levels, a most specific domain level having a highest degree of specificity may be identified. The term may be associated with a domain-specific term associated with the term at the most specific domain level. The domain-specific term associated with the term may be stored.

Abstract: The disclosed subject matter generally relates to a voice assistant system for a vehicle which can distinguish who is issuing voice commands, and thereby also which sub-system in the vehicle to control according to the voice command. This is obtained by allowing the voice assistant system to receive input signals from an image capturing system configured to detect the user issuing the voice command. The vehicle compartment may be divided into several zones each associated with at least one sub-system, and the input signals from the image capturing system provide information to the control unit of the voice assistant system regarding the present zone of the user issuing the voice commands.

Abstract: A computing system identifies an incoming voice call from a user device to an agent device associated with the computing system. The computing system generates a transcription of the incoming voice call using one or more natural language processing techniques. The computing system extracts a problem description from the transcription. The problem description indicates a topic for the incoming voice call. A first machine learning model estimates a situation vector from the problem description. A second machine learning model identifies a pre-existing situation vector that closely matches the estimated situation vector. The computing system retrieves a situation description that corresponds to the identified pre-existing situation vector.

Abstract: An apparatus has a network interface circuit to receive a source sentence from a network connected client device. A processor is connected to the network interface circuit. A memory is connected to the processor. The memory stores parameters of a multilingual neural review system and instructions executed by the processor to operate the multilingual neural review system trained on a corpus of source sentences, draft target sentences and corrected target sentences. The multilingual neural review system produces a corrected target sentence from a draft target sentence representing a proposed translation of the source sentence. The draft target sentence and the corrected target sentence are supplied to the network connected client device. Approval for the corrected target sentence is received from the network connected client device. Parameters of the multilingual neural review system are updated based upon the approval for the corrected target sentence.