Abstract: Embodiments of the disclosed technologies use machine learning to produce thread level classification data and case level classification data.

Abstract: Techniques for hybrid behavioral pairing in a contact center system are disclosed. In one embodiment, the techniques may be realized as a method for hybrid behavioral pairing in a contact center system comprising: determining a first ordering of a plurality of contacts according to a behavioral pairing strategy with a balanced contact utilization; determining a second ordering of the plurality of contacts according to a performance-based routing strategy with an unbalanced contact utilization; determining a third ordering of the plurality of agents according to a combination of the first ordering and the second ordering having a skewed contact utilization between the balanced contact utilization and the unbalanced contact utilization; and outputting a hybrid behavioral pairing model based on the third ordering for connecting an agent to a contact of the plurality of contacts in the contact center system.

Abstract: A system and method for mobile device active callback integration, utilizing a callback integration engine operating on a user's mobile device that present a callback token for integration through the operating system and software applications operating on the device, wherein interacting with the callback token produces a callback object used to execute a callback incorporating device hardware, context, scheduling, and trust information.

Abstract: A device-management system performs processing, such as audio processing, in an instance of a virtual machine corresponding to a functionally limited (local) device. To register the local device, the device-management system receives a registration request that includes device information, encryption data, and an indication of an associated user account. The device-management system then sends this registration data to a service-provider system, which returns a shared encryption key. The device-management system and the local device may use this shared encryption key to securely communicate. The device-management system may de-allocate the instance upon detecting a period of inactivity of the local device and may re-allocate the instance when new activity is detected. The device-management system may further determine when and if audio data to be sent to the local device is encoded using a codec not implemented by the local device.

Abstract: In a method of configuring a private branch exchange remotely, a user transmits configuration data to a local private branch exchange via a callable device. The local private branch exchange transmits the configuration data to a remote private branch exchange. Thereby, the user can conveniently configure the private branch exchange configuration or the private branch exchange service of the remote private branch exchange. Besides, the method of configuring a private branch exchange remotely can be applied to private branch exchanges manufactured by different manufacturers as well as private branch exchanges managed by different enterprises or organizations.

Abstract: A system and method for creating organized intent clusters or scenes using machine learning algorithms is provided. A method of creating organized intent clusters or scenes comprises extracting intent features related to the plurality of request inputs. The method also includes creating a plurality of groups comprising the extracted intent features. The method includes identifying a cluster based on co-occurring extracted intent features, the co-occurring extracted intent features belonging to a plurality of domains. The method further includes generating a proto-scene based in part by ranking the extracted intent features within the cluster.

Abstract: Methods and systems for training a language processing model. The methods may involve receiving a first log record in a first format, wherein the first log record includes annotations describing items in the first log record, and then creating a second log record in a second format comprising data from the first log record utilizing the annotations in the first log record and a conversion rule set. The second log record may then be used to train a language processing model so that a trained model can identify items in a third log record and the relationships therebetween.

Abstract: System and method for handling a transaction between a waiting party and queuing party include an independent communication system (ICS) managing calls between the waiting party and calling party for handling sensitive data as well as call-attached data. The ICS manages the transaction in different stages and with different levels of sensitivity. Either party is allowed to modify the call or call preferences during the transaction. The ICS works independently from the queuing party calling system.

Abstract: Techniques for behavioral pairing in a contact center system are disclosed. In one particular embodiment, the techniques may be realized as a method for pairing in a contact center including ordering one or more contacts, ordering one or more agents, comparing a first difference in ordering between a first contact and a first agent in a first pair with a second difference in ordering between a second contact and a second agent in a second pair, and selecting the first pair or the second pair for connection based on the comparing, wherein the first contact and the second contact are different or the first agent and the second agent are different.

Abstract: A method to allow for real-time dictation and transcription with multiple remote endpoints is provided. The method comprises evoking a primary application and a client device APP (or APP) to work with a remote hosted application to process audio for the primary application. The APP connects to the hosted application, and the hosted application receives and processes the audio. The hosted application returns the text to the client device, which text populates the primary application. The APP and/or the hosted application also transmits the text to a remote endpoint, such as, for example, a desktop computer or a laptop computer where the user can interact with the primary application and the text returned by the hosted application.

Abstract: A method of determining a highlight section of a sound source includes obtaining a sound source and classification information of the sound source, and learning a neural network by using the sound source and the classification information. The neural network includes an input layer including a node corresponding to a feature value of each of a plurality of sections obtained by splitting the sound source according to a time axis, an output layer including a node corresponding to the classification information, a hidden layer defined between the input layer and the output layer, a first function between the input layer and the hidden layer, and a second function between the hidden layer and the output layer, wherein the first function includes an attention model for calculating a weighted sum of the feature value of each section. The highlight section of the sound source is determined based on weight information of a feature value node of each section included in the first function.

Abstract: This application discloses methods and systems for semantic classification of numerical data in a natural language context. The methods and systems employ a machine learning model with a convolutional neural network as a feature detector and a feedforward neural network as a numerical data classifier.

Abstract: A method including receiving, in a machine learning model (MLM), a corpus including words. The MLM includes layers configured to extract keywords from the corpus, plus a retrospective layer. A first keyword and a second keyword from the corpus are identified in the layers. The first and second keywords are assigned first and second probabilities. Each probability is a likelihood that a keyword is to be included in a key phrase. A determination is made, in the retrospective layer, of a first probability modifier that modifies the first probability based on a first dependence relationship between the second keyword being placed after the first keyword. The first probability is modified using the first probability modifier. The first modified probability is used to determine whether the first keyword and the second keyword together form the key phrase. The key phrase is stored in a non-transitory computer readable storage medium.

Abstract: Techniques for hybrid behavioral pairing in a contact center system are disclosed. In one embodiment, the techniques may be realized as a method for hybrid behavioral pairing in a contact center system comprising: determining a first ordering of a plurality of agents according to a behavioral pairing strategy with a balanced agent utilization; determining a second ordering of the plurality of agents according to a performance-based routing strategy with an unbalanced agent utilization; determining a third ordering of the plurality of agents according to a combination of the first ordering and the second ordering having a skewed agent utilization between the balanced agent utilization and the unbalanced agent utilization; and outputting a hybrid behavioral pairing model based on the third ordering for connecting a contact to an agent of the plurality of agents in the contact center system.

Abstract: A process for responding to location-variable group electronic digital assistant requests includes monitoring audio communications transmitted between a plurality of communication devices (CDs) on a group communication channel (CC) and detecting an audio inquiry from a first CD of the plurality of CDs. A location of each CD is determined, including different first and second determined locations of the first and second CDs. The audio inquiry is processed and it is determined that a response will vary based on the different first and second locations. Two different first and second responses to the audio inquiry are generated as a function of the determined locations and are caused to be provide to respective ones of the first and second CDs via one of (i) two private call CCs, (ii) one individual private call CC and the group CC, and (iii) the group CC and a second group CC.

Abstract: There are provided a method, a system configured to perform the method, and a computer program product comprising instructions which when executed by processing circuitry of the system cause the system to perform the method. The method comprises determining, by at least one computer processor communicatively coupled to and configured to operate in a contact center system, a plurality of available contacts, wherein each of the plurality of available contacts is available for pairing to an agent of a plurality of agents using either a first pairing strategy or a second pairing strategy, wherein the first pairing strategy is different from the second pairing strategy.

Abstract: Method and apparatus to analyze and present location information in an easy-to-digest manner are disclosed. In one embodiment, each piece of location information can include a piece of location-designating information and a piece of location-related information. Location-designating information is primarily for identifying location. Location-related information is information related to location-designating information. The location-designating information and the location-related information can be supplied by a mobile device. With the help of location-related information, each piece of location-designating information can be more accurately transformed into a label to help identify a location. The amount of location information can be reduced. All of the location-designating information pertaining to a given area can be consolidated into one piece of location-designating information related to the label.

Abstract: One variation of method includes, at a mobile radio: receiving an audio communication from an originating device over a wireless network at a first time; storing the audio communication in local memory; outputting the audio communication, at approximately the first time, at a first volume level set manually by a user associated with the mobile radio prior to the first time; calculating a confidence score that the audio communication was comprehended by the user based on the first volume setting; accessing a priority of the audio communication defined at the originating device; and, in response to a discrepancy between the confidence score and the priority of the audio communication, replaying the audio communication, from local memory, at a second volume setting greater than the first volume setting.

Abstract: A method of enhancing distorted signal having a speech and noise component, with a processing device having memory with stored training information T, comprising a step of removing noise in spectrum domain according to a noise and speech model to obtain a clear signal spectrum, wherein the training information T comprises dictionaries of compact spectra prototypes of speech and noise, speech gains and noise gains forming together composite states and it comprises probabilities of state history. Dictionaries of compact spectra prototypes comprise compact spectra prototypes of a first dimension P1 and of a second dimension P2 where second dimension P2 is higher or equal to the first dimension P1.

Abstract: A computer-implemented learning method includes combining a first score for a word included in a dictionary of a model and a second score for the word calculated based on a length of the word and a number of remaining characters up to an upper limit number of characters of a summary, the first score being output in response to input of input text to the model, calculating, based on a combined score generated by the combining, a distribution of generation probabilities of the word, and updating, based on the calculated distribution of the generation probabilities, parameters in the model.