Customer contact service with real-time supervisor assistance

Patent metadata

Field	Value
Publication number	US-11893526-B2
Application number	US-201916698470-A
Country	US
Kind code	B2
Filing date	Nov 27, 2019
Priority date	Nov 27, 2019
Publication date	Feb 6, 2024
Grant date	Feb 6, 2024

Abstract

Systems and methods to implement customer contact service with real-time supervisor assistance. A supervisor may oversee multiple agents in a customer contact service. A service of a computing resource service provider may monitor a plurality of audio connections at a service of a computing resource service provider, generate transcripts for the plurality of audio data, analyze the transcripts using a set of natural language processing (NLP) techniques to generate metadata, tag the transcripts with categories based at least in prat on the metadata, generate information for at least a portion of the plurality of connections based on the transcripts, metadata, and categories, and provide the information to a supervisor of the agents.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
connecting a first service to a plurality of audio streams of calls between agents and customers;
obtaining a plurality of audio data for the calls at the first service;
using a second service to generate transcripts for a plurality of audio data, wherein a turn of the transcripts is reconstructed based at least in part on a first fragment of the transcripts and a second fragment of the transcripts that correspond to each of the plurality of audio that that was obtained at a different time;
as a result of determining that the turn of the transcripts is reconstructed, analyzing the transcripts with a third service to generate a set of natural language processing (NLP) outputs encoding audio characteristics associated with the transcripts;
tagging the transcripts with categories based at least in part on the set of NLP outputs, wherein the categories are defined based at least in part on rules that evaluate content and audio characteristics, and at least one of the rules is determined by a supervisor of the agents;
generating a notification for the plurality of audio streams based on the categories, where the notification identifies a new pattern from at least two of the plurality of audio streams during a specified period; and
providing the notification to the supervisor.

2. The computer-implemented method of claim 1 , wherein Web Socket connections are used to connect the first service with the plurality of audio streams.

3. The computer-implemented method of claim 1 , wherein the set of NLP outputs includes an output determined by performing sentiment analysis, entity detection, or key phrase detection.

4. The computer-implemented method of claim 1 , wherein the providing the notification to the supervisor includes presenting the notification to the supervisor on a graphical dashboard, sending a text message to the supervisor, or sending an email to the supervisor.

5. A system, comprising:
one or more processors, and
memory that stores computer-executable instructions that, if executed, cause the system to:
establish a first communications channel between a service of a computing resource service provider and a first client of the computing resource service provider, wherein the first client is connected to an entity via a second communications channel;
obtain, at the service and over the first communications channel, audio data of the second communications channel;
generate a transcript based at least in part on the audio data, wherein a portion of the transcript is reconstructed based at least in part on subportions of the transcript that correspond to separate portions of the audio data obtained at a different time;
as a result of reconstructing the portion of the transcript, generate, using a set of natural language processing (NLP) techniques, metadata encoding audio characteristics associated with the transcript;
tag the transcript with one or more categories based at least in part on the set of NLP outputs, wherein the one or more categories are defined based at least in part on rules that evaluate content and audio characteristics, and at least one of the rules is obtained by a second client of the computer resource service provider; and
provide, to the second client, a notification that:
indicates that the second communications channel was tagged with the one or more categories; and
indicates that a new issue from the one or more categories occurred within a specified period.

6. The system of claim 5 , wherein the first communications channel is a Web Socket connection and the second communications channel is a phone call.

7. The system of claim 5 , wherein the set of NLP techniques includes: p sentiment analysis, entity detection, or key phrase detection.

8. The system of claim 7 , wherein a category of the one or more categories identifies a negative trending sentiment of the entity on the second communications channel.

9. The system of claim 5 , wherein a category of the one of the categories identifies a manner in which the first client is to interact with the entity on the second communications channel.

10. The system of claim 5 , wherein the instructions to generate the transcript based at least in part on the audio data include instructions to:
buffer at least a portion of the audio data that includes a first portion of a turn;
obtain, at the service and over the first communications channel, additional audio data of the second communications channel, the additional audio including a second portion of the turn; and
generate the transcript based at least in part on the audio data and the additional audio data.

11. The system of claim 5 , wherein the notification includes aggregate information from a plurality of connections between other clients and other entities.

12. The system of claim 11 , wherein the aggregate information includes a common issue identified from at least a portion of the audio data, and the common issue is distinct from the new issue.

13. A non-transitory computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to at least:
establish a plurality of communications channels between a service of a computing resource service provider and a plurality of clients of the computing resource service provider;
obtain a plurality of contacts data from the plurality of communications channels; generate transcripts for the plurality of contacts data, wherein a portion of the transcripts is reconstructed based at least in part on at least two subportions of the transcripts that correspond to separate portions of the plurality of contacts data obtained at a different time;
as a result of reconstructing the portion of the transcripts, generate, using a set of natural language processing (NLP) techniques, metadata encoding conversation characteristics associated with the transcripts;
tag the transcript with one or more categories based at least in part on the metadata, wherein the one or more categories are defined based at least in part on rules that evaluate content and conversation characteristics, and at least one of the rules is determined by an entity that manages the plurality of clients;
generate information for the plurality of communications channels based on the transcripts, metadata, and categories, where the information comprises a new trend identified from at least two of the plurality of contacts data during a specified period; and
provide the information to the entity.

14. The non-transitory computer-readable storage medium of claim 13 , wherein the plurality of communications channels are real-time communications channels.

15. The non-transitory computer-readable storage medium of claim 13 , wherein the instructions to generate, using the set of natural language processing (NLP) techniques, metadata encoding the conversation characteristics associated with the transcripts include instructions to submit a request to a second service of the computing resource service provider to run the set of NLP techniques.

16. The non-transitory computer-readable storage medium of claim 13 , wherein the one or more categories are based on one or more audio characteristics of the contacts data.

17. The non-transitory computer-readable storage medium of claim 13 , characteristics wherein the one or more conversation characteristics are one or more audio comprising: speaking volume, periods of silence, or interruptions.

18. T

Assignees

• Amazon Tech Inc

Inventors

• Sivasubramanian Swaminathan
• Philomin Vasanth
• Anbazhagan Vikram
• Singh Ashish
• Deo Atul
• Arora Anuroop
• Davidson Colin Thomas
• Young Jessie
• El-Haggan Yasser

Classifications

• G06Q10/0633 Primary

  Workflow analysis · CPC title

• G06F16/685

  using automatically derived transcript of audio data, e.g. lyrics (speech recognition G10L15/00) · CPC title

• G06F40/16

  Automatic learning of transformation rules, e.g. from examples · CPC title

• G10L15/1815

  Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning · CPC title

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title