In-call virtual assistant
An in-call virtual assistant system monitors a real-time call, e.g., a call that is in progress, between multiple speakers, identifies a trigger and executes a specified task in response to the trigger. The virtual assistant system can be invoked by an explicit trigger or an implicit trigger. For example, an explicit trigger can be a voice command from one of the speakers in the call, such as “Ok Chorus, summarize the call” for summarizing the call. An implicit trigger can be an event that occurred in the call, or outside of the call but that is relevant to a speaker. For example, an event such as a speaker dropping off the call suddenly can be an implicit trigger that invokes the virtual assistant system to perform an associated task, such as notifying the remaining speakers on the call that one of the speakers dropped.
1. A computer-implemented method, comprising:
receiving real-time call data of a call that is in progress between a representative and a customer;
extracting multiple features from the real-time call data, wherein the multiple features include any of characteristics of the representative, characteristics of the customer and characteristics associated with a conversation of the call;
monitoring the real-time call data to identify a trigger for invoking a virtual assistant component, wherein the virtual assistant component executes a specified task associated with the call;
receiving the trigger to invoke the virtual assistant component; and
executing, by the virtual assistant component and using the multiple features, the specified task.
2. The computer-implemented method of claim 1 further comprising:
generating a result of the specified task, and
generating a notification regarding the result of the specified task for at least one of the representative and customer.
3. The computer-implemented method of claim 2 , wherein generating the notification includes generating at least one of a textual notification or an audio notification.
4. The computer-implemented method of claim 1 , wherein receiving the trigger includes receiving a voice command from the representative for invoking the virtual assistant component.
5. The computer-implemented method of claim 4 , wherein receiving the voice command includes receiving a specified word or a specified phrase from the representative that is explicitly indicative of invoking the virtual assistant component.
6. The computer-implemented method of claim 4 , wherein receiving the voice command includes receiving the specified task from the representative.
7. The computer-implemented method of claim 1 , wherein receiving the trigger includes determining an event associated with the call that is implicitly indicative of invoking the virtual assistant component.
8. The computer-implemented method of claim 7 , wherein determining the event includes determining a first feature of the multiple features that is implicitly indicative of invoking the virtual assistant component.
9. The computer-implemented method of claim 8 , wherein determining the first feature includes determining a set of words or phrases in the call that implicitly invokes the virtual assistant component.
10. The computer-implemented method of claim 8 , wherein determining the first feature includes determining external data associated with the call that is implicitly indicative of invoking the virtual assistant component.
11. The computer-implemented method of claim 10 , wherein determining the external data includes determining information associated with the customer based on metadata of the call.
12. The computer-implemented method of claim 10 , wherein determining the external data includes determining information associated with the representative based on metadata of the call.
13. The computer-implemented method of claim 10 , wherein determining the external data includes determining information associated with the conversation based on metadata of the call.
14. The computer-implemented method of claim 10 , wherein determining the external data includes obtaining the external data from one or more computer servers accessible by the virtual assistant component using metadata of the call.
15. The computer-implemented method of claim 1 , wherein executing the specified task includes generating a summary of the conversation based on the multiple features.
16. The computer-implemented method of claim 15 , wherein generating the summary includes generating the summary by performing a semantic analysis of the multiple features, the multiple features including one or more words or phrases in the call.
17. The computer-implemented method of claim 15 , wherein generating the summary includes generating the summary using any of rule-based, artificial intelligence (AI), machine-learning (ML), or natural language processing (NLP) techniques.
18. The computer-implemented method of claim 1 , wherein executing the specified task includes generating an action item of the call based on the multiple features, the multiple features including one or more words or phrases in the call that are indicative of the action item.
19. The computer-implemented method of claim 1 , wherein extracting the multiple features includes extracting the multiple features using any of AI, NL, or NLP techniques.
20. The computer-implemented method of claim 1 , wherein the call is at least one of a video call, an online meeting, a virtual reality-based call, an augmented reality-based call, or an email interaction between the customer and the representative.
21. The computer-implemented method of claim 1 , wherein receiving the trigger includes receiving a series of voice commands from the representative, wherein a second command of the series of voice commands is received after the completion of a task of a first command of the series of voice commands by the virtual assistant component.
22. A non-transitory computer-readable storage medium storing computer-readable instructions, comprising:
instructions for receiving real-time call data of a call that is in progress between multiple speakers;
instructions for extracting multiple features from the real-time call data, wherein the multiple features include at least one of characteristics of a first speaker of the multiple speakers and characteristics associated with a conversation of the call;
instructions for receiving, by a virtual assistant component, a voice command for executing a specified task associated with the call; and
instructions for executing, by the virtual assistant component and using the multiple features, the specified task.
23. The non-transitory computer-readable storage medium of claim 22 , wherein the instructions for executing the specified task include:
instructions for analyzing the multiple features to obtain data for performing the specified task, and
instructions for executing the specified task based on the data.
24. The non-transitory computer-readable storage medium of claim 23 , wherein the instructions for analyzing the multiple features include:
instructions for obtaining the data from analyzing the multiple features, deriving the data from metadata of the call, or deriving the data based on data obtained from analyzing the multiple features.
25. The non-transitory computer-readable storage medium of claim 22 , wherein the instructions for receiving the voice command include:
instructions for receiving one or more words from the first speaker for invoking the virtual assistant component.
26. The non-transitory computer-readable storage medium of claim 25 , wherein the instructions for receiving the voice command include:
instructions for receiving a set of words from the first speaker that is indicative of the specified task.
27. The non-transitory computer-readable storage medium of claim 22 further comprising:
instructions for determining an event associated with the call that is implicitly indicative of invoking the virtual assistant component.
28. The non-transitory computer-readable storage medium of claim 27 , wherein the instructions for determining the event include:
instructions for determining a first feature of the multiple features that is implicitly indicative of invoking the virtual assistant component, the first feature including one or more words from the call that implicitly invokes the virtual assistant component.
29. The non-transitory computer-readable storage medium of claim 27 , wherein the instructions for determining the event include:
instructions for determining a first feature of the multiple features that is indicative of an event external to the call but associated with the first speaker that is implicitly indicative of invoking the virtual assistant component.
30. The non-transitory computer-readable storage medium of claim 29 , wherein the instructions for determining the first feature further include:
instructions for analyzing, by the virtual assistant component, the first feature to determine the specified task, and
instructions for executing the specified task by the virtual assistant component.
31. A system, comprising:
a processor;
a first component that is configured to receive real-time call data of a call that is in progress between a representative and a customer;
a second component that is configured to extract multiple features from the real-time call data, wherein the multiple features include any of characteristics of the representative, characteristics of the customer and characteristics associated with a conversation of the call; and
a third component that is configured to monitor the real-time call data for a trigger to invoke a virtual assistant for executing a specified task, and to invoke the virtual assistant; and
a fourth component that is configured to execute, using the multiple features, the specified task.