摘要:
A method for creating a textual summary of a call includes transcribing speech to text in real time using a speech-to-text generating unit configured for execution upon one or more data processors, automatically matching, in real-time, text to predetermined intents and extracted entities using an intent recognizing unit for execution upon the one or more data processors, automatically mapping the predetermined intents and extracted entities into a call summary using one or more mapping functions, and displaying the call summary using an agent user interface for execution upon the one or more data processors. A contact center call summarization system may include a contact center communication device, a speech-to-text generating unit, an intent recognizing unit, and an agent user interface.
摘要:
Systems and methods are described herein for providing a Voice over Internet Protocol (VoW) call. In an embodiment, a load balancing processor receives a re-initiated HTTP request from a client processor upon detection that an initial call server is no longer active, and sends the re-initiated HTTP request to a second call server. The second server generates updated call resource information that identifies the second server as the new server resource for the call, and sends the updated call resource information over the IP network to the client processor. Subsequent HTTP requests from the client processor for sending and receiving signaling and media data for the call are received at the second server using the updated call resource information.
摘要:
A method for determining a number of calls to generate for a specific outbound campaign in an automated contact center based upon a prediction of agent occupancy in a future time period, wherein the automated contact center processes one or more outbound campaigns. The estimates a total number of outbound agents available to the one or more outbound campaigns. By using the total number of outbound agents, an effective number of agents for the specific outbound campaign is determined according to a ratio. The effective number of agents is used to determine a number of outbound calls to generate for the specific outbound campaign according to a dialing method. Calls are then generated for the specific outbound campaign based upon the determined number of outbound calls to generate.
摘要:
A method for determining a number of calls to generate in an automated contact center based upon a prediction of agent occupancy in a future time period. The automated contact center comprises a dialer for dialing generated calls, a queue where successfully connected dialed calls await agent handling, and an agent pool where calls are handled by agents. The method estimates a number of calls presently in the dialer, the queue, and the agent pool that would be in the agent pool in the future time period. Based upon the estimations, the method predicts a future agent occupancy in the future time period, which is used to determine the number of calls to generate in the automated contact center. Calls are generated based upon the number of calls to generate.
摘要:
A method for determining a number of calls to generate in an automated contact center, wherein the automated contact center comprises a dialer for dialing generated calls and an agent pool where successfully connected dialed calls are handled by agents. The method collects empirical data on durations in which calls spend in the dialer and agent handling times. The collected data is used to generate discrete density or discrete distribution functions. Using the discrete density or distribution functions, an individual impact is calculated for each call presently in the dialer. The method uses the individual impacts to calculate a dialer impact, which is used to predict a future agent occupancy. Based upon the prediction, a number of calls to generate is determined. Calls are generated based upon the number of calls to generate.
摘要:
Systems and methods are provided for training natural language processing (NLP) models in a contact center. The systems and methods provided may render the process of training an NLP model faster, easier to perform, and more accessible to non-experts. In embodiments of the present disclosure, a method for training an NLP model includes a first step of initializing an NLP model with an intent and one or more seed phrases. The next step may be to receive a customer interaction. Next, a matched utterance can be generated based on the customer interaction and the NLP model. Then, a suggested training phrase may be generated based on the matched utterance. The suggested training phrase may be confirmed. Thereafter, the NLP model can be updated with the confirmed training phrase.
摘要:
A system for social care routing, prioritization and agent assistance for interfacing with a carrier network, a social listening device, a social network or proprietary social feedback apparatus, and a contact center and contact center database. Author postings, demographic information, sentiment, topical relevancy, and customer service requests and other data are used for prioritization, agent assistance and routing. A contact center database and a social network listening device or proprietary social feedback apparatus obtain information used in determining routing and tagging instructions. A user interface is connected to the system to accept configurable conditions for determining customer service instructions to the agents. A color-coded agent heads-up display for author and customer profiling and customer relationship management timeline is disclosed for effectively managing social posts and authors needing customer service assistance by agents at each target enterprise contact center.
摘要:
A method for controlling and correcting abandonment rate in an automated contact center that uses a predictive dialing method for determining a number of calls to generate for dialing. When abandonment rate is close to zero, the method determines a mean agent occupancy that is used by the predictive dialing method. Empirical data is collected on agent occupancy while the predictive dialing method uses the mean agent occupancy to determine a number of calls to generate. The method determines a target abandonment rate and a target agent occupancy, wherein a proportion of the agent occupancy empirical data having an agent occupancy equal to or higher than the target agent occupancy is within a predetermined range of the target abandonment rate. The target agent occupancy is used to adjust the mean agent occupancy used by the predictive dialing method.
摘要:
A method for creating a textual summary of a call includes transcribing speech to text in real time using a speech-to-text generating unit configured for execution upon one or more data processors, automatically matching, in real-time, text to predetermined intents and extracted entities using an intent recognizing unit for execution upon the one or more data processors, automatically mapping the predetermined intents and extracted entities into a call summary using one or more mapping functions, and displaying the call summary using an agent user interface for execution upon the one or more data processors. A contact center call summarization system may include a contact center communication device, a speech-to-text generating unit, an intent recognizing unit, and an agent user interface.
摘要:
Systems and methods are described herein for providing a Voice over Internet Protocol (VoIP) call. In an embodiment, a load balancing processor receives a re-initiated HTTP request from a client processor upon detection that an initial call server is no longer active, and sends the re-initiated HTTP request to a second call server. The second server generates updated call resource information that identifies the second server as the new server resource for the call, and sends the updated call resource information over the IP network to the client processor. Subsequent HTTP requests from the client processor for sending and receiving signaling and media data for the call are received at the second server using the updated call resource information.