Real-time linguistic and acoustic detection for context-rich notifications and coaching
A real-time contextual event notification system ingests events as streams from any authorized entity, applies rules to the event streams, determines a context of an end-user who is a recipient of a targeted notification, and provides notifications to the end-user in accordance with the context. The event streams may come from multiple sources, and rules may be applied to provide real-time contextual information associated with the end user. One such event stream includes detected linguistic and/or acoustic events during a phone call between two or more persons.
1 . A system for detecting audio events during an audio interaction between a user and a customer, the system comprising:
one or more processors; and
memory having instructions stored thereon that, when executed by the one or more processors, cause the system to:
analyze audio data during the audio interaction to detect an audio event, wherein the audio event includes a linguistic event and an acoustic event, wherein the acoustic event comprises a sound from the audio interaction other than a spoken word;
enrich the audio event with contextual data from one or more additional data streams responsive to the detection of the audio event, wherein the contextual data indicates a length of the audio interaction, a current status of the audio interaction, and a status of the customer, and wherein the contextual data includes application event messages received from a remote device over a socket connection;
generate a notification based on the enriched audio event and the current status of the audio interaction during the audio interaction, wherein the notification includes an indication of the audio event and an indication of the contextual data that was used to enrich the audio event; and
present the notification to the user via a user interface.
2 . The system of claim 1 , wherein the audio data is analyzed and the audio event enriched in real-time or near real-time.
3 . The system of claim 1 , wherein the notification includes an indication of the audio event and the contextual data.
4 . The system of claim 1 , wherein the current status of the audio interaction comprises one of whether or not the customer is on hold and whether or not the audio interaction is being recorded.
5 . The system of claim 4 , wherein enriching the audio event with contextual further comprises merging the contextual data with the audio event.
6 . The system of claim 4 , wherein the instructions further cause the system to:
authenticate the application event messages and the audio data using a token that is contained within a header prior to analyzing the audio data.
7 . The system of claim 1 , wherein the instructions further cause the system to:
determine the current status of the audio interaction based on the contextual data; and
continuously update a cache based on the current status of the audio interaction.
8 . The system of claim 1 , wherein the notification includes at least one interactive icon, wherein the instructions further cause the system to:
receive a user input via the at least one interactive icon; and
initiate a response action responsive to the user input.
9 . The system of claim 1 , wherein the notification includes one or more response actions for the user.
10 . A method of providing real-time contextual event notifications to a user during an audio interaction between the user and a customer, the method comprising:
receiving, by an API gateway component, event messages from a source over a socket connection, the event messages comprising information relating to a linguistic event, an acoustic event, and an application event, wherein the acoustic event comprises a sound from the audio interaction other than a spoken word;
authenticating, by the API gateway component, the event messages using a token that is contained within a header;
processing, by a managed streaming component, the event messages to determine a context to generate event notifications, wherein the context is determined based on the linguistic event, the acoustic event, and the application event, wherein the context includes a length of the audio interaction, a current status of the audio interaction, and a status of the customer, and wherein the contextual data includes application event messages received from a remote device over a socket connection; and
communicating event notifications to a client management service (CMS) that determines a target client to receive the event notifications, wherein the event notifications includes an indication of the audio interaction and an indication of the contextual data.
11 . The method of claim 10 , wherein the audio interaction is a phone call and wherein linguistic events or acoustic events are detected by analyzing the phone call in real-time.
12 . The method of claim 10 , wherein processing the event messages further comprises:
determining a current state of the audio interaction associated with the event messages;
updating a cache with a most current event of the audio interaction; and
adding the event notifications for consumption by the CMS for delivery to the target client.
13 . The method of claim 10 , further comprising receiving feedback on the event notifications from the target client.
14 . The method of claim 10 , further comprising:
receiving a connection from the target client; and
sending scripting language code to the target client that defines a user interface.
15 . The method of claim 10 , further comprising:
receiving the event messages as parallel streams;
processing the parallel streams by topic; and
determining the context in real-time as the parallel streams are received.
16 . A non-transitory computer readable medium comprising instructions that, when executed by one or more processors of a computing device, cause the computing device to:
analyze audio data during an audio interaction between a user and a customer to detect an audio event, wherein the audio event includes at least one of a linguistic event and an acoustic event, wherein the acoustic event comprises a sound from the audio interaction other than a spoken word;
enrich the audio event with contextual data from one or more additional data streams responsive to the detection of the audio event, wherein the contextual data indicates a length of the audio interaction, a current status of the audio interaction, and a status of the customer, and wherein the contextual data includes application event messages received from a remote device over a socket connection;
generate a notification based on the enriched audio event and the current status of the audio interaction during the audio interaction, wherein the notification includes an indication of the audio event and an indication of the contextual data that was used to enrich the audio event; and
present the notification to the user via a user interface.
17 . The computer readable medium of claim 16 , wherein the notification includes an indication of the audio event and the contextual data.
18 . The computer readable medium of claim 16 , wherein the current status of the audio interaction comprises one of whether or not the customer is on hold and whether or not the audio interaction is being recorded.
19 . The computer readable medium of claim 16 , wherein enriching the audio event with contextual further comprises merging the contextual data with the audio event.
20 . The computer readable medium of claim 16 , wherein the instructions further cause the computing device to:
authenticate application event messages and the audio data using a token that is contained within a header prior to analyzing the audio data.