1166results about "Automatic exchanges" patented technology

A generative mediator engine can perform a requested interaction with a conversational agent of a target entity on behalf of a user. An internal conversational platform can identify intents for the requested interaction. An external artificial intelligence engine can perform intent discovery when an intent is not identified above a confidence threshold. A discovered intent unknown to the generative mediator engine can be received from the external artificial intelligence engine and used, with input requirements determined by the generative mediator for the requested interaction, by a dialog generator to generate a sample dialog for the requested interaction. User feedback can be received after review of action items and expected inputs identified from the sample dialog. The generative mediator engine can perform the requested interaction with the conversational agent on behalf of the user and without receiving user intervention during the requested interaction.

A method can include receiving, from a customer, a request for a support encounter. A method can include determining a customer intent indicative of an issue experienced by the customer. A method can include routing the customer to a support agent based on the intent. A method can include generating, using a first large language model, summaries of previous support encounters. A method can include providing the summaries to the support agent. A method can include determining, based on the intent and / or the summaries, one or more support documents related to the issue. A method can include generating, using a second large language model, a summary of each of the support documents, wherein an input to the second large language model comprises a support document. A method can include providing the generated summaries of the support documents to the support agent for use in mitigating the issue.

A computing system displays, via a first display generation component, a computer-generated environment, and concurrently displays status information corresponding to the computing system via the second display generation component, including concurrently displaying: a visual representation of a portion of a user of the computing system who is in a position to view the computer-generated environment via the first display generation component, and one or more graphical elements that provide a visual indication of content in the computer-generated environment. The computing system detects a respective event; and in response, the computing system changes a level of immersion of the computer-generated environment displayed via the first display generation component, and changes the status information that is displayed via the second display generation component, including changing an appearance of the visual representation of the portion of the user of the computing system.

A system for an automated real-time fraud detection based on predictive analytics of call data including a processor of a call analysis server (CAS) node configured to host a machine learning (ML) module and connected to at least one user-entity node over a network and a memory on which are stored machine-readable instructions that when executed by the processor, cause the processor to: capture user call data including a plurality of key elements from the at least one user-entity node; parse the user call data to extract a plurality of classifying features based on the plurality of key elements; query a local calls database to retrieve local historical calls-related data based on the plurality of classifying features; generate at least one feature vector based on the plurality of classifying features and the local historical calls-related data; provide the at least one feature vector to the ML module coupled to an Artificial Neural Network (ANN); receive a plurality of user call ranking parameters from a call predictive model generated by the ML module using outputs of the ANN based on the feature vector; and generate a fraud verdict for the user call based on the plurality of call ranking parameters.

An autonomous communication system includes a language model, a retrieval module, a caching mechanism, an autonomous agent, and a human interface for managing interactions and responses. A method and computer-readable medium for managing communication also include these components for processing, enhancing, summarizing, managing interactions, and allowing human intervention.

Disclosed is technology that analyzes voicemail data and takes a particular action based on the analysis of the data. For example, a user device (e.g., a smart phone) can receive a voicemail from an external caller and analyze data representative of the voicemail with an artificial intelligence agent. The artificial intelligence agent can utilize contextual data including user device contextual data and past voicemail contextual data to determine the appropriate task to perform based on the voicemail. Based on that determination, the user device can automatically perform the task that is relevant to the voicemail on behalf of the user.

A method, system, and article of manufacture for modifying a queue assignment for a contact center are provided. At least one computing device determines, by solving for constraints, at least one staffing change for at least one of the first queue or the second queue, using as inputs (a) first time-based service metrics, (b) second time-based service metrics, (c) data describing eligible staffing changes, and (d) at least one predetermined constraint. The at least one staffing change is a modification to the queue assignment of a subset of the computing device instances that is predicted to improve net staffing relating to service targets selected from (a) an average speed of answer (ASA) specifying an average time for one of the computing device instances to initiate handling of a particular task after the particular task has been assigned to the relevant queue and (b) a service level specifying a percentage of tasks for which handling is initiated within a specified time frame after the tasks are assigned to the queue. The at least one staffing change can be implemented by modifying the queue assignment of the subset of the computing device instances.

A system of a telecommunications network that uses IP Multimedia Subsystem (IMS) network elements to screen inbound calls. A Media Resource Function (MRF) intercepts an inbound call and uses various techniques to screen the call. The MRF redirects the call to an interactive voice response (IVR) system, which prompts the caller to state the caller's name, records the caller's response, and relays it to the called subscriber, who can then decide to accept or reject the call. The system can present a random number challenge to the caller. The system can use machine learning analytics and a large language model (LLM) to analyze the response's contents, accepting the call only if it relates to a topic allowed by the subscriber, perform audio analysis to determine whether the caller sounds robotic, or detect whether the response contains other known robocall markers.

The invention discloses a power customer service real-time assistance and work order linkage method and system, and the method comprises the steps: collecting the voice data of an incoming call in real time, translating the voice data in real time to generate a translated text, and synchronously extracting the voice emotion features; performing semantic recognition on the translated text to extract a business appeal, performing risk analysis through a multi-dimensional risk early warning algorithm in combination with the translated text and the voice emotion features, and outputting a risk level; automatically judging and marking customer types and risk labels based on the translated text and the risk level; and automatically generating and filling a work order according to the business appeal, the customer type and the risk tag, and associating the work order with the translated text and the voice data. Through deep fusion of voice real-time translation and risk early warning, the problems of high translation delay, incomplete risk early warning coverage and low system cooperation efficiency in the prior art are solved, full-process automatic closed-loop management is realized, and the service efficiency, the risk management and control capability and the customer satisfaction are remarkably improved.

The invention relates to the technical field of communication, in particular to a dynamic service scheduling method for a 5G new call based on intelligent semantics, which realizes efficient scheduling of 5G call services through three core steps of multi-operator fusion access, dynamic routing distribution and intelligent voice interaction. According to the method, a unified SIP trunk line integration platform is constructed, and network resources of three major operators are integrated; respectively designing routing strategies for low-concurrency and high-concurrency scenes from the dimensions of network quality, cost rate, compliance, link load and the like, and dynamically optimizing path selection through a weight algorithm; and intelligent analysis and interaction of voice semantics are realized by combining ASR, TTS and NLP engines. According to the method, the problems of cross-operator routing optimization, high-concurrency processing and compliance management are solved, and the method is suitable for scenes with high requirements on stability and compliance, such as finance and government affairs.

The invention relates to the technical field of artificial intelligence, and particularly discloses an artificial intelligence-based automatic identification method and system for fraud-related numbers, which breaks through the limitation that a traditional method depends on single-dimensional data through a data acquisition module, comprehensively describes a number behavior mode through multi-source data fusion, reduces erroneous judgment and missed judgment caused by one-sided data, and improves the identification accuracy of fraud-related numbers. A data foundation is laid for accurate identification; the feature library construction module introduces a knowledge graph technology, constructs the number, the associated object and the behavior event into a graph, and combines data cleaning and feature extraction and screening to form a dynamically updated fraud-related feature library; the method not only captures the surface behavior characteristics of the number, such as the call duration and the short message content, but also can mine the hidden fraud relation network to update the novel fraud characteristics in time, thereby solving the problem that the traditional method lags in the recognition of the complex fraud mode, and improving the timeliness and depth of the feature library.

A computer-implemented machine learning method for improving speaker separation is provided. The method comprises processing audio data to generate prepared audio data and determining feature data and speaker data from the prepared audio data through a clustering iteration to generate an audio file. The method further comprises re-segmenting the audio file to generate a speaker segment and causing to display the speaker segment through a client device.

Systems and methods for orchestrating interaction with an artificial intelligence (AI) application in a contact center environment receive, via an AI agent, a voice message from a user; convert the message from voice to text; generate an initial computational inference process based on the text message; determine whether or not all information required to execute the initial computational inference process is available to the processor; when a determination is made that all information required is available: execute the initial computational inference process; generate a text reply based on the initial computational inference process; convert the text reply to a voice reply; and send the voice reply to the user via the AI agent; when a determination is made that information is unavailable: generate a text query requesting the information; convert the text query to a voice query; and send the voice query to the user via the AI agent.

A method is provided that is performed at a system including multiple speech collectors to collect speech from a talker to produce corresponding ones of multiple audio signals that each convey speech energy: for each audio signal: separating high-frequency speech energy from low-frequency speech energy; and determining a first energy level of the high-frequency speech energy; and determining a preferred audio signal among the multiple audio signals for subsequent processing at least based on the first energy level of each audio signal.

Aspects of the present disclosure are directed to receiving a user phone call and triggering user interactions via an alternative channel in response to the user phone call. Users may call an account service provider with issues related to a user account. To service such a call, the account service provider may prompt the user for information. This information can be analyzed to determine whether the conditions for the user's call are aligned with a shift to a more efficient alternative channel. For example, a guided user workflow via a webpage, application, chat agent, etc. may represent a more efficient deployment of organizational resources. When such conditions are aligned, the account service provider may trigger a selected user interaction workflow for the user call via an alternative channel. The user interaction workflow can comprise a guided workflow that is configured to resolve the user's issue related to the call.

The invention relates to the technical field of audio processing, in particular to an airborne high-reliability intelligent voice call squelch reduction method, which comprises the following steps: performing ADC conversion on an acquired analog audio signal to obtain a digital audio signal; carrying out framing and FFT conversion on the digital audio signal, carrying out phase optimization, calculating a dynamic spectrum entropy weight, combining an auditory masking compensation factor, carrying out spectrum synthesis, and carrying out IFFT conversion and overlap elimination to obtain an audio signal after sound mixing processing; noise reduction processing is carried out on the audio signal after sound mixing processing through multi-mode feature extraction, and an audio signal after noise reduction processing is obtained; performing squelch judgment of Bayesian decision optimization on the audio signal subjected to noise reduction processing through energy feature extraction to obtain an audio signal subjected to squelch processing; performing automatic gain processing on the audio signal after the squelch processing to obtain an audio signal after automatic gain; and performing DAC conversion on the audio signal after automatic gain, and outputting a noise-reduced analog audio signal.

Systems and methods for providing contextual based actions based on a natural language input are disclosed. The method comprises: receiving, on a media device, a natural language input; determining, based on the natural language input, a first context of the natural language input; and determining, based on the first context, a first action. a dynamic action button is generated and configured to be selected by a user to carry out an action, and in response to the user selecting the dynamic action button, the systems and methods describe carrying out the first action.

Based on an initial set of claim data for a claim event of a user, a computing system can generate an initial script and a content flow that includes the initial script for a call representative to utilize in a call session with the user, the content flow comprising a plurality of sub-flow groupings. Based on the user initiating a topic corresponding to a respective sub-flow grouping from the plurality of sub-flow groupings, the system can update the content flow to enable the call representative to hop to the respective sub-flow grouping within the content flow.

Systems and methods for facilitating electronic safety alert communications by a safety alert management system are disclosed herein. A method includes receiving an electronic safety alert for a specific safety event from a user electronic device via a safety alert application, the safety alert including at least one user message from a user associated with the user device. The method includes initiating an electronic chat session between the user and the safety agent attending the safety management application and, for at least one user message received at the safety management application, determining a reply message to send to the user device in response to the at least one user message. In embodiments, a machine learning model is used to analyze the user message and determine one or more recommended reply messages to display to the agent. A method for training a safety chat language model in a safety alert management system is also disclosed.

According to one aspect of the present disclosure, the voice message analysis system receives a voice message from a voicemail system, performs a voice-to-text transcription, and applies a set of rules to the transcription of the voice message. The actions have associated data that is used when implementing the action. The rules specify criteria for triggering various actions that map to functions of the voicemail system or other functions. The actions have associated data that is used in implementing the actions. A voice message analysis system applies the voice message analysis rules to the transcript of the voice message. Based on a determination that the voice message matches a voice message analysis rule, the voice message system implements the appropriate action, such as by programmatically calling the appropriate function of the voicemail system and providing data for the function based on the data associated with the first action.

The invention discloses a self-adaptive dialogue generation method based on a personalized memory map and a related device, and the method comprises the steps: enabling a long-term memory database on a far-end server to be associated with an intelligent outbound robot all-in-one machine after logging in the far-end server; loading the personalized memory map to the local of the intelligent outbound robot all-in-one machine for local storage; the method comprises the following steps: when entering a chat conversation mode, collecting conversation audio data of a service object user, and extracting audio emotion information, semantic information and semantic keywords in the conversation audio data; matching knowledge nodes of the semantic keywords in the personalized memory graph; and dialogue audio data corresponding to the dialogue audio data is generated based on the semantic information, the knowledge nodes and the audio emotion information, and dialogue playing processing is carried out. In the embodiment of the invention, the lightweight local storage of the intelligent outbound robot all-in-one machine is realized, the conversation with the user is realized through the personalized memory map, and the sense of reality is improved.

The embodiment of the specification provides a gas operation method and an Internet of Things system based on a smart gas call center. The method comprises the following steps: acquiring gas user call data of the call center; predicting demand information of different types of users based on the gas user call data, wherein the demand information at least comprises gas product demand and gas service demand; determining and pushing gas operation push features based on the demand information of different types of users, wherein the gas operation push features comprise push type features and push content features.

Systems and methods detect and prevent vishing attacks through an integrated framework combining SIP header customization, STIR / SHAKEN frameworks, AI / ML analysis, and real-time speech analysis using the Viterbi algorithm. The system begins with call initiation, embedding authentication information in the SIP header. The SIP data is transmitted and verified using STIR / SHAKEN frameworks, ensuring the authenticity of the caller's identity. Verified data is cross-referenced with third-party databases and analyzed by an AI / ML engine to detect anomalies. If potential fraud is detected, the call is blocked, and the customer is notified. Calls that pass initial checks are further analyzed using the Viterbi algorithm, which converts speech to text and identifies suspicious patterns. An anomaly pattern detector processes the converted text to detect vishing indicators, terminating the call if a match is found. This multi-layered approach ensures robust protection against vishing, enhancing the security and reliability of voice communications while safeguarding users from fraud.

A system, device, and method to reconnect a terminated communication to the original queue and provide queue status over a different communication channel. The method includes receiving a first communication over a first communication channel and placing the first communication in a contact center queue. The method also includes providing a user with an option to terminate the first communication over the first communication channel and receive queue status information over a second communication channel, and in response to the user selecting the option: terminating the first communication while maintaining a queue status in the contact center queue for the first communication; providing the queue status information; in response to the queue status of the first communication being within a threshold, notifying the user to initiate a second communication over the first communication channel; and replacing the first communication in the contact center queue with the second communication.

A telecommunications switch-type platform positioned between a private-branch exchange and a data network dynamically hosts artificial-intelligence agents that interpret, transcribe, translate and transliterate live call traffic while the session is in progress. Machine-readable instructions stored in local memory instantiate, migrate and terminate the agents on demand across dedicated processing resources such as field-programmable gate arrays, application-specific integrated circuits or system-on-chip devices, thereby maintaining end-to-end latency below 250 milliseconds. An integrated call-audio recorder captures bidirectional media streams for secure archiving without interrupting service. The platform exposes a network interface that passes both signalling and media traffic, enabling seamless deployment inside call-centre or enterprise environments and compatibility with softswitch architectures conforming to class-4 Computational Interpretation of Communication using Artificial Intelligence standards. The same functional stack is deliverable as a computer-implemented method and as a non-transitory machine-readable medium storing the instructions executed by the processing resources.

The present disclosure provides a system for adaptive voicemail and interactive voice response (IVR) detection in outbound calls. The system includes a call initialization module configured to establish an outbound call connection, a speech processing module configured to convert incoming audio signals into text in real-time, a classification module configured to analyze the text and determine whether the call has reached a live recipient, a voicemail system, or an IVR menu, a decision-making module configured to determine an appropriate course of action based on the classification, and a response generation module configured to generate and deliver appropriate responses based on the determined course of action. The system enables efficient handling of outbound calls by accurately detecting and responding to different call scenarios.

The invention discloses an intelligent voice outbound system capable of realizing automatic adjustment, and relates to the technical field of intelligent voice interaction, the system comprises a voice acquisition and preprocessing module, a voiceprint analysis and emotion recognition module, a dynamic verbal skill generation module, a verbal skill real-time optimization module and a voice synthesis and outbound module; the communication efficiency and personalized service are improved through a'generation-optimization-feedback 'closed-loop design, the system generates scene-fitting verbal skills according to customer emotions, historical records and demands, adjusts strategies according to instant feedback of customers, reduces invalid conversations, reduces customer conflicts, shortens the communication period, and meanwhile, improves the user experience. Voiceprint features are decomposed into four quantifiable dimensions, emotion is accurately recognized in combination with a combination rule, limitation of a single index is avoided, the dynamic verbal skill generation module is based on a generative adversarial network, high-quality verbal skill can be learned, the dynamic verbal skill generation module adapts to multiple scenes, customization cost is reduced, and intelligent outbound large-scale application is promoted.

A method implemented by a system, the method comprising: receiving a call; determining whether an automatic number identification of the call is valid; and routing the call to an agent.

The invention relates to the technical field of intelligent outbound calls, in particular to an intelligent outbound call processing method and device based on artificial intelligence and a server. Real-time voice data input by a user is converted into text data based on a voice analysis model, intention analysis is conducted on the text data based on an intention analysis model, and the intention of the user is obtained. And performing sentiment analysis on the text data based on the sentiment analysis model to obtain a sentiment analysis result of the user, generating a dialogue strategy matched with the user according to a user intention analysis result and the sentiment analysis result, and finally executing dynamic adjustment operation according to the dialogue strategy and dialogue skill data to obtain verbal skill data matched with the user. According to the method, the accurate dialogue strategy is generated for the user not only through the intention of the user but also in combination with the sentiment analysis result of the user, so that the verbal skill data to be provided for the user is adjusted, the determination accuracy and precision of the verbal skill data needed by the user are improved, and the interaction accuracy of the intelligent outbound dialogue process is improved.

A system and method for multi-channel dynamic advertisement testing. The system comprises a multi-platform adaptive ad campaign manager, a dynamic advertisement engine, a campaign database, and an omnichannel text-based communicator. The system receives customer interactions with two advertisement test variants, establishes a real-time media stream between a customer device and a second user device, and monitors the media stream to collect data related to effectiveness of the advertisement variants. The system may analyze media stream data together with a plurality of other data types to statistically determine which of the two advertisement variants resulted in better performance based on a variety of advertisement metrics. The system may use the plurality of data and the statistical analysis to suggest an advertisement element to be altered in the next round of advertisement variant testing. This system can combine data collection and analytics for an ad campaign together into one system.