U.S. patent application number 16/186502 was filed with the patent office on 2020-05-14 for caller deflection and response system and method.
The applicant listed for this patent is Nuance Communications, Inc.. Invention is credited to Theodore Dougherty, Matthew E. Ellis, Adam Mak, Adam Stuczynski.
Application Number | 20200153965 16/186502 |
Document ID | / |
Family ID | 70550927 |
Filed Date | 2020-05-14 |
United States Patent
Application |
20200153965 |
Kind Code |
A1 |
Dougherty; Theodore ; et
al. |
May 14, 2020 |
CALLER DEFLECTION AND RESPONSE SYSTEM AND METHOD
Abstract
Provided are a call deflection and response system and method,
wherein a voice call from a caller device is received, a skill
group is determined to resolve an issue associated with the call,
and a text response to the issue is sent to the caller device,
providing a context-based personalized response. A caller leaves a
detailed voicemail explaining an issue needing resolution, which is
electronically transcribed and then run through a classifier to
determine concepts and intents associated with the call. Based on
the concepts and intents, responsibility for the call and
associated files are transferred to a particular skill group on a
chat platform for resolution. A chat entity from the appropriate
skill group determines and provides an issue response via text
message to the caller device, e.g., to the caller's mobile
phone.
Inventors: |
Dougherty; Theodore;
(Denver, CO) ; Mak; Adam; (Etobicoke, CA) ;
Stuczynski; Adam; (Winnetka, CA) ; Ellis; Matthew
E.; (Dobbs Ferry, NY) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Nuance Communications, Inc. |
Burlington |
MA |
US |
|
|
Family ID: |
70550927 |
Appl. No.: |
16/186502 |
Filed: |
November 10, 2018 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04M 2201/60 20130101;
H04M 2201/41 20130101; H04M 2201/40 20130101; H04M 2203/6054
20130101; H04M 3/5233 20130101; H04L 51/02 20130101; H04M 3/5166
20130101; H04L 51/04 20130101; H04M 3/5183 20130101; H04M 2203/6045
20130101; H04M 3/42382 20130101 |
International
Class: |
H04M 3/51 20060101
H04M003/51; H04L 12/58 20060101 H04L012/58; H04M 3/523 20060101
H04M003/523 |
Claims
1. A method, carried out by a plurality of processors accessing one
or more computer memory, the method comprising: electronically
transcribing at least one audio file to form an electronic
transcription, the at least one audio file associated with an audio
call describing an issue; processing the electronic transcription
to determine concepts and intents of the audio call, the concepts
and intents indicating a subject matter of the issue; determining a
skill group from a plurality of skill groups based on the
determined concepts and intents, the skill group having a
predetermined association with the subject matter; and
electronically forwarding the transcription to a response system
including the skill group for determining an issue response.
2. The method of claim 1, further comprising electronically
forwarding the at least one audio file and/or a transcription
confidence score to the response system.
3. The method of claim 1, further comprising electronically
authenticating a caller and/or a caller device associated with the
audio call.
4. The method of claim 3, wherein electronically authenticating the
caller and/or the caller device includes using voice biometrics
authentication processing.
5. The method of claim 3, further comprising: electronically
accessing stored account information associated with the
authenticated caller and/or caller device; and electronically
forwarding the stored account information to the response
system.
6. The method of claim 1, further comprising an interactive voice
response (IVR) system applying natural language processing and/or
conversational artificial intelligence to engage in a dialog during
an audio call with a caller device to obtain the at least one audio
file.
7. The method of claim 6, further comprising, before determining
the issue response: the IVR system terminating the audio call; and
a text processor sending an acknowledgment text message to the
electronic device.
8. The method of claim 1, wherein transcribing the at least one
audio file includes processing the at least one audio file using
natural language processing.
9. The method of claim 1, further comprising: automatically
prepopulating fields in an electronic preformatted response
template with data based on the transcription; and electronically
forwarding, the electronic response template to the response
system.
10. The method of claim 1, wherein the method further comprises
generating a text message and/or a chat message comprising the
issue response and transmitting the text message and/or the chat
message to an electronic device.
11. The method of claim 10, wherein the text message and/or the
chat message is generated based on a populated response
template.
12. The method of claim 1, further comprising the response system
electronically accessing one or more knowledge bases or systems to
determine the issue response.
13. The method of claim 1, further comprising receiving audio
content from a mobile device via the audio call, storing the audio
content as the at least one audio file, terminating the audio call
with the mobile device, and electronically communicating an issue
response to the mobile device.
14. A call processing system, comprising: a transcription processor
configured to electronically transcribe at least one audio file to
form an electronic transcription, the at least one audio file
associated with an audio call describing an issue; a classifier
configured to; analyze the electronic transcription to determine
concepts and intents of the audio call, the concepts and intents
indicating a subject matter of the issue; and determine a skill
group from a plurality of skill groups based on the determined
concepts and intents, the skill group having a predetermined
association with the subject matter; and a deflection processor
configured to electronically forward the transcription to a
response system including the skill group for determination of an
issue resolution.
15. The system of claim 14, wherein the deflection processor is
further configured to electronically forward the at least one audio
file and/or a transcription confidence score to the response
system.
16. The system of claim 14, further comprising an authentication
processor configured to electronically authenticate a caller and/or
a caller device associated with the audio call.
17. The system of claim 16, wherein the authentication processor is
configured to electronically authenticate the caller and/or the
caller device using voice biometrics authentication processing.
18. The system of claim 16, wherein: the authentication processor
is configured to electronically access stored account information
associated with the authenticated caller and/or caller device; and
the deflection processor is configured to electronically forward
the stored account information to the response system,
19. The system of claim 14, further comprising; an interactive
voice response (IVR) system configured to apply natural language
processing and/or conversational artificial intelligence to engage
in a dialog with a caller device to obtain the at least one audio
file.
20. The system of claim 19, wherein the IVR system is configured to
terminate the audio call and the deflection and response system
further comprise a text processor configured to send an
acknowledgment text message to the electronic device in response to
termination of the call.
21. The system of claim 14, wherein the transcription processor is
configured to transcribe the at least one audio file includes
processing the at least one audio file using natural language
processing.
22. The system of claim 14, wherein the system is configured to:
automatically prepopulate fields in an electronic preformatted
response template with data based on the transcription; and
electronically forward the electronic response template to the
response system.
23. The system of claim 14, wherein the response system is
configured to generate a text message and/or a chat message
comprising the issue response and to transmit the text message
and/or the chat message to an electronic device.
24. The system of claim 23, wherein the response system is
configured to generate the text message and/or the chat message
based on a populated response template.
25. The system of claim 14, wherein the response system is
configured to electronically access one or more knowledge bases or
systems to determine the issue response.
26. The system of claim 14, wherein the system is configured to
receive audio content from a mobile device via the audio call,
store the audio content as the at least one audio file, terminate
the audio call with the mobile device, and electronically
communicate an issue response to the mobile device,
27. The system of claim 26, wherein the issue response comprises a
text message.
28. The system of claim 27, wherein the text message comprises
alphanumeric words, phrases, numbers, icons, graphics, audio,
video, hyperlinks, or a combination of two or more thereof.
29. The method of claim 13, wherein the issue response comprises a
text message.
30. The method of claim 29, wherein the text message comprises
alphanumeric words, phrases, numbers, icons, graphics, audio,
video, hyperlinks, or a combination of two or more thereof.
Description
FIELD OF INTEREST
[0001] The present inventive concepts relate to the field of call
response systems, and more particularly to customer call response
systems using interactive voice response (IVR).
BACKGROUND
[0002] Customer service phone calls are frequently handled, at
least initially, by an interactive voice response (IVR) system.
Most digital transformation initiatives at enterprises feature a
requirement to deflect and offload phone calls to agents to
lower-cost digital channels. This often includes an offer to engage
the caller with a chat agent who will open up a chat session with
the caller, forcing that caller to start the resolution process
over.
[0003] For instance, in existing IVR to SMS deployments, the caller
is offered an opportunity to chat with an agent. If the caller opts
for that option, the caller will receive an offer to chat via an
SMS, but the chat agent has no other information about the caller.
They are starting from scratch, even though the caller may have
gone through numerous IVR prompts, has waited on hold, and
typically has given some amount of information to the IVR by virtue
of spoken or DTMF-based response.
[0004] That is, the new resolution process forces the customer to
engage in an often lengthy communication process, with numerous
back and forth text messages required to resolve an issue. As more
and more calls are deflected in this manner, this creates the
potential to create another bad customer experience in the Short
Message Service (SMS) texting channel, as has happened in the
interactive voice response (IVR) voice channel.
SUMMARY
[0005] Provided are a deflection and response system and method. In
accordance with the inventive concepts, such systems and methods
deflect voice calls received from an IVR system to a deflection and
response system having text capability. The deflection and response
system is configured to resolve a caller's issue by replying with a
context-based personalized response in the form of a text message,
e.g., a SMS/MMS/RCS text message. The system enables the method by
allowing individuals to state an issue to be resolved, i.e., the
reason for the call. The caller's audio inputs are recorded and
processed to determine a caller issue from information determined,
at least in part, from the audio input. The system further enables
the method by determining an issue resolution, and sending a
response associated with the resolution via a text message to the
caller's electronic device (e.g., mobile phone). The text message
could include alphanumeric words, phrases, numbers, icons,
graphics, audio, video, hyperlinks, and the like, and/or
combinations of two or more thereof. The intent can be a request
for a problem solution and/or information, as examples.
[0006] Systems and methods in accordance with the inventive
concepts provide, therefore, an improved caller (or customer)
experience, with service delivered at a far lower cost than
provided by existing approaches. Each successfully deflected call
yields an immediate financial benefit in the form of lower cost of
transaction, and potentially a mid- and long-term benefit for an
entity servicing the caller by moving the caller (or customer) from
a high-cost service channel (live voice) to a lower cost digital
channel (chat) on an ongoing basis.
[0007] In accordance with aspects of the inventive concepts,
provided is a call deflection and response method, carried out by a
plurality of processors accessing one or more computer memory. The
method comprises electronically transcribing at least one audio
file to form an electronic transcription, the at least one audio
file associated with an audio call describing an issue; processing
the electronic transcription to determine concepts and intents of
the audio call; determining a skill group from a plurality of skill
groups based on the determined concepts and intents, the concepts
and intents indicating a subject matter of the issue and the skill
group having a predetermined association with the subject matter;
and electronically forwarding the transcription to a response
system including the skill group for determining an issue
response.
[0008] In various embodiments, the method further comprises
electronically forwarding the at least one audio file and/or a
transcription confidence score to the response system.
[0009] In various embodiments, the method further comprises
electronically authenticating a caller and/or a caller device
associated with the audio call.
[0010] In various embodiments, electronically authenticating the
caller and/or the caller device includes using voice biometrics
authentication processing.
[0011] In various embodiments, the method further comprises
electronically accessing stored account information associated with
the authenticated caller and/or caller device and electronically
forwarding the stored account information to the response
system.
[0012] In various embodiments, the method further comprises an
interactive voice response (IVR) system applying natural language
processing and/or conversational artificial intelligence to engage
in a dialog during an audio call with a caller device to obtain the
at least one audio file.
[0013] In various embodiments, the method further comprises, before
determining the issue response the IVR system terminating the audio
call and a text processor sending an acknowledgment text message to
the electronic device.
[0014] In various embodiments, transcribing the at least one audio
file includes processing the at least one audio file using natural
language processing.
[0015] In various embodiments, the method further comprises
automatically prepopulating fields in an electronic preformatted
response template with data based on the transcription and
electronically forwarding the electronic response template to the
response system.
[0016] In various embodiments, the response system is a chat
platform and the method further comprises generating a text and/or
chat message comprising the issue response and transmitting the
text and/or chat message to an electronic device.
[0017] In various embodiments, the text and/or chat message is
generated based on a populated response template.
[0018] In various embodiments, the method further comprises the
response system electronically accessing one or more knowledge
bases or systems to determine the issue response.
[0019] In various embodiments, the method further comprises
receiving audio content from a mobile device via the audio call,
storing the audio content as the at least one audio file,
terminating the audio call with the mobile device, and texting an
issue response to the mobile device.
[0020] In accordance with another aspect of the inventive concepts,
provided is a deflection and response system. The system comprises
a transcription processor configured to electronically transcribe
at least one audio file to form an electronic transcription, the at
least one audio file associated with an audio call describing an
issue. The system also includes a classifier configured to analyze
the electronic transcription to determine concepts and intents of
the audio call, the concepts and intents indicating a subject
matter of the issue, and to determine a skill group from a
plurality of skill groups based on the determined concepts and
intents, the skill group having a predetermined association with
the subject matter. And the system includes a deflection processor
configured to electronically forward the transcription to a
response system including the skill group for determination of an
issue resolution.
[0021] In various embodiments, the deflection processor is further
configured to electronically forward the at least one audio file
and/or a transcription confidence score to the response system.
[0022] In various embodiments, the system further comprises an
authentication processor configured to electronically authenticate
a caller and/or a caller device associated with the audio call.
[0023] In various embodiments, the authentication processor is
configured to electronically authenticate the caller and/or the
caller device using voice biometrics authentication processing.
[0024] In various embodiments, the authentication processor is
configured to electronically access stored account information
associated with the authenticated caller and/or caller device and
the deflection processor is configured to electronically forward
the stored account information to the response system.
[0025] In various embodiments, the system further comprises an
interactive voice response (IVR) system configured to apply natural
language processing and/or conversational artificial intelligence
to engage in a dialog with a caller device to obtain the at least
one audio file.
[0026] In various embodiments, the IVR system is configured to
terminate the audio call and the deflection and response system
further comprise a text processor configured to send an
acknowledgment text message to the electronic device in response to
termination of the call.
[0027] In various embodiments, the transcription processor is
configured to transcribe the at least one audio file includes
processing the at least one audio file using natural language
processing.
[0028] In various embodiments, the system is configured to
automatically prepopulate fields in an electronic preformatted
response template with data based on the transcription and
electronically forward the electronic response template to the
response system.
[0029] In various embodiments, the response system is a chat
platform configured to generate a text and/or chat message
comprising the issue response and to transmit the text and/or chat
message to an electronic device.
[0030] In various embodiments, the chat platform is configured to
generate the text and/or chat message based on a populated response
template.
[0031] In various embodiments, the response system is configured to
electronically access one or more knowledge bases or systems to
determine the issue response.
[0032] In various embodiments, the system is configured to receive
audio content from a mobile device via the audio call, store the
audio content as the at least one audio file, terminate the audio
call with the mobile device, and text an issue response to the
mobile device.
BRIEF DESCRIPTION OF THE DRAWINGS
[0033] The present invention will become more apparent in view of
the attached drawings and accompanying detailed description. The
embodiments depicted therein are provided by way of example, not by
way of limitation, wherein like reference numerals refer to the
same or similar elements. The drawings are not necessarily to
scale, emphasis instead being placed upon illustrating aspects of
the invention. In the drawings:
[0034] FIG. 1 is flow diagram representing an embodiment of aspects
of a call deflection and response method, in accordance with
aspects of the inventive concepts;
[0035] FIG. 2 is an embodiment of a more detailed call deflection
and response method, in accordance with aspects of the inventive
concepts;
[0036] FIG. 3 is an embodiment of a call deflection and response
system, in accordance with aspects of the inventive concepts;
[0037] FIG. 4 is a diagram depicting an embodiment of a caller's
audio interaction with the deflection and response system of FIG.
3, in accordance with aspects of the inventive concepts; and
[0038] FIGS. 5A through 5C depict an embodiment of text message
interactions between the caller's electronic device and the
deflection and response system of FIG. 3, in accordance with
aspects of the inventive concepts.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
[0039] Various aspects of the inventive concepts will be described
more fully hereinafter with reference to the accompanying drawings,
in which some exemplary embodiments are shown. The present
inventive concept may, however, be embodied in many different forms
and should not be construed as limited to the exemplary embodiments
set forth herein.
[0040] It will be understood that, although the terms first,
second, etc. may be used herein to describe various elements, these
elements should not be limited by these terms. These terms are used
to distinguish one element from another, but not to imply a
required sequence of elements. For example, a first element can be
termed a second element, and, similarly, a second element can be
termed a first element, without departing from the scope of the
present invention. As used herein, the term "and/or" includes any
and all combinations of one or more of the associated listed items.
The term "or" is not used in an exclusive or sense, but in an
inclusive or sense.
[0041] It will be understood that when an element is referred to as
being "on" or "connected" or "coupled" to another element, it can
be directly on or connected or coupled to the other element or
intervening elements can be present. In contrast, when an element
is referred to as being "directly on" or "directly connected" or
"directly coupled" to another element, there are no intervening
elements present. Other words used to describe the relationship
between elements should be interpreted in a like fashion (e.g.,
"between" versus "directly between," "adjacent" versus "directly
adjacent," etc.).
[0042] The terminology used herein is for the purpose of describing
particular embodiments only and is not intended to be limiting of
the invention. As used herein, the singular forms "a," "an" and
"the" are intended to include the plural forms as well, unless the
context clearly indicates otherwise. It will be further understood
that the terms "comprises," "comprising," "includes" and/or
"including," when used herein, specify the presence of stated
features, steps, operations, elements, and/or components, but do
not preclude the presence or addition of one or more other
features, steps, operations, elements, components, and/or groups
thereof.
[0043] To the extent that functional features, operations, and/or
steps are described herein, or otherwise understood to be included
within various embodiments of the inventive concept, such
functional features, operations, and/or steps can be embodied in
functional blocks, units, modules, operations and/or methods. And
to the extent that such functional blocks, units, modules,
operations and/or methods include computer program code, such
computer program code can be stored in a computer readable medium,
e.g., such as non-transitory memory and media, that is executable
by at least one computer processor.
[0044] Provided are a call deflection and response system and
method. In accordance with the inventive concepts, such systems and
methods can receive voice calls from an interactive automated
response system and determine an issue resolution and reply to the
caller with a context-based personalized response in the form of a
text message, e.g., a personalized SMS/MMS/RCS/iMessage text
message. In accordance with aspects of the inventive concepts, the
system and method improve upon existing approaches by allowing
callers to leave a detailed reason for their call, e.g., as a
voicemail, and later receive a text response with an issue
resolution. The deflection and response system can utilize a chat
platform, e.g., to a live chat agent or chatbot (collectively, a
"chat entity"), to more quickly resolve the caller's issue. A
"chatbot" is computer program that, when executed, simulates a text
conversation with human users, while a chat agent tends to be a
human. The resolution is provided via text message to the caller,
e.g., to the caller's mobile phone. The caller can, if necessary,
respond to the text message and engage in a chat session to further
pursue the issue, if necessary.
[0045] FIG. 1 is flow diagram 100 representing an embodiment of
aspects of the deflection and response method, in accordance with
aspects of the inventive concepts. FIG. 3 provides an embodiment of
a deflection and response system 350. In block 110, the deflection
and response system 350 receives a call from a caller device 10 and
records audio from the caller e.g., which can take the form of a
voicemail left through interaction with an IVR system 356. In block
120, the recorded audio files are transcribed into text files. In
block 130, the transcribed text files are analyzed by a classifier
to determine concepts and intents of the caller and, based thereon,
responsibility for the call is selectively assigned and routed to
an appropriate skill group, e.g., from among a plurality of skill
groups. Each skill group can have a predetermined association with
a subject matter, e.g., billing, technical support, cable service,
Internet service, and so on. Based on the concepts and intents, a
subject matter and skill group can be determined, e.g., using
natural language processing to understand concepts and intents in
the caller's audio file. Within a skill group, there can be one or
more specialists equipped to determine an issue resolution. The
skill group is preferably selected, therefore, based on the
concepts and intents of the caller. Specialists from each skill
group preferably have access to systems-of-record and resources
needed to enable research and resolution of the caller's issue.
That is, from the recorded and transcribed audio and determined
concepts and intents, one or more caller issues can be determined
requiring resolution, e.g., billing questions, service
interruption, and/or service modification.
[0046] In block 140, a specialist from the selected skill group
researches the issue and determines a resolution. In block 150, the
specialist completes the resolution process by replying to the
caller's device with a personalized context-based
SMS/MMS/RCS/iMessage or other text message, which can include a
free form answer and one-time rich media link, as examples.
[0047] FIG. 2 is a flowchart of a more detailed embodiment of a
call deflection and response method 200, in accordance with aspects
of the inventive concepts.
[0048] In step 210, an automated call system 310 receives a call
(see arrow 1) from a caller device 10, e.g., a mobile phone. The
automated call system 310 can be configured to service at least one
business, governmental, charitable, or other organization, and can
form part of the enterprise system 200. The automated call system
310 can include an interactive voice response (IVR) system 312
configured to interact with the caller using a caller device 10,
such as a mobile phone, computer, tablet, phablet, or other call
device. Although a mobile phone 10 is depicted, is not at the
exclusion of other electronic communication devices or systems.
Interaction between the automated call system 310 and the caller
(via caller device 10) enables the automated call system 310 to
receive information from the caller device 10, which can include
audio inputs, keypad entries, a phone number or other
identification of the caller device 10, and so on.
[0049] Optionally, the automated call system 310 can be configured
to provide some level of authentication using backend systems 320,
which can include an authentication processor, a customer
relationship management (CRM) system, and databases or knowledge
bases 322, and/or other systems and databases having information
and resources useful or necessary for obtaining the call
information.
[0050] In response to inputs from the caller device 10, the
automated call system 310 determines a destination for the call to
one or more possible exit points. In particular, the automated call
system 310 can be configured with capabilities to forward
responsibility for the call to one or more exit points via an exit
point router 330. The exit points can include the deflection and
response system 350 and, in some embodiments, one or more other
possible exit points, such as a call center 340 with live and/or
automated call agents. In step 220, the automated call system 310
directs the call to the deflection and response system 350 via the
exit point router 330.
[0051] In various embodiments, the automated call system 310 can be
configured to automatically transfer responsibility for the call to
the deflection and response system 350 (as an exit point) in
response to satisfaction of one or more criteria or the existence
of one or more conditions. As examples, such criteria and/or
conditions can include one or more predetermined criteria and/or
conditions, such as a wait time estimated at or above a threshold
as calculated by a wait time counter, an elapsed amount of wait
time (e.g., an elapsed amount of time at or beyond a present
threshold time limit) as calculated by the wait time counter,
and/or an identity of the caller, a caller account, and/or the
caller device 10. Other conditions could additionally or
alternatively exist for automatically initiating call transfer to
the deflection and response system 350 as an exit point.
[0052] In various embodiments, the automated call system 310 can be
configured to allow the caller to select or initiate transfer of
responsibility for the call to the deflection and response system
350, e.g., by pressing a keypad button or submitting an appropriate
and/or predetermined audio response via the caller device 10. In
such embodiments, the automated call system 310 could be configured
to solicit an input from the caller device 10 to affect a call
transfer to the deflection and response system 350 via the exit
point router 330.
[0053] The deflection and response system 350 can be local or
remote to other components of an enterprise system, such as an
enterprise system that includes the automated call system 310
and/or the backend systems 320. In various embodiments, the
automated call system 310 can be configured to transfer
responsibility for the call to the deflection and response system
350 by one or more computer transfer mechanisms (e.g., TNT/Transfer
Connect, SIP Refer, etc.).
[0054] In step 230, the deflection and response system 350 receives
the transferred call from the exit point router 330. The deflection
and response system 350 includes a deflection processor 352
responsible for overall management and tasking of the deflection
and response system 350 and its components. The deflection
processor 352 tasks an IVR system 356 to take over the call and
engage with caller device 10 (see arrow 2) to obtain call
information. The IVR system 356 can be, in some embodiments, a
Conversational Interactive Voice Response IVR by Nuance
Communications.
[0055] In various embodiments, the deflection and response system
350 optionally includes an authentication processor 353 configured
to authenticate the caller, the caller device 10, and/or a caller
account, in step 233. The IVR system 356 can be configured to
engage with the caller to obtain information necessary or useful by
the authentication processor 353. In some embodiments, the
authentication processor 353 can interact with the various backend
systems 320, as needed, for the authentication process.
Authenticated caller's data can be written to an encrypted folder
and passed as electronic files via a web service to the deflection
and response system 350, which can be adapted to comply with
enterprise security requirements.
[0056] In some embodiments, the authentication processor 353 can
use voice biometrics technology to authenticate the caller, e.g.,
by sampling enough audio data in the caller's voice interaction
with the IVR system 356 to perform the authentication. Any now
known or later developed voice biometrics authentication system
could be used in various embodiments. This can include leveraging
previously recorded and processed speech by the caller and
repurposing prior voice data to improve caller authentication.
[0057] In some embodiments, the authentication processor 353 can
include voice biometrics authentication processing, e.g., such as
provided by VocalPassword.TM. and/or FreeSpeech.TM., both presently
offered by Nuance Communications, Inc. VocalPassword is a biometric
authentication system that authenticates a caller when the caller
speaks a simple passphrase to validate its identity. And FreeSpeech
is a biometric authentication system that authenticates a caller by
listening to their normal conversation with the IVR, where the
input of passwords and/or PINs is not required. The dialog example
in FIG. 4 demonstrates interactions between the IVR system 356 and
caller device 10 where the authentication processor 353 is
invoked.
[0058] In some embodiments, other forms of authentication can be
used, e.g., such as techniques that use a password, a personal
identification number (PIN), and/or other personally identifiable
inputs.
[0059] The IVR system 356 can be configured to apply scripts,
natural language processing, and/or conversational artificial
intelligence to engage in a conversation with the caller via caller
device 10 to obtain the call information (e.g., see FIG. 4),
including for the optional authentication processing discussed
above. The IVR system 356 can engage with the caller device 10 to
populate pre-existing call templates ultimately useful by a skill
group for resolving the caller's issue.
[0060] In some embodiments, the IVR system 356 can interact with a
wait time counter to determine an estimated wait time for
availability of a live agent, and can indicate the estimated wait
time to the caller as part of its caller interaction. The IVR
system 356 can obtain call information and invite the caller to
leave a voicemail describing the issue, which can form at least
part of the call information. Therefore, the call information can
include caller audio information, which can be recorded by the IVR
system 356, in step 230. The call information can also include a
caller identity and/or caller account information. The call
information can be stored as electronic information, which can
include a phone number and account information associated with the
caller and/or the caller device 10 as well as the recorded audio.
Some or all of the files can be encrypted.
[0061] Once the caller has left its voicemail and it has been
recorded, the IVR system 356 can terminate the call with the caller
device 10, in step 235. The deflection and response system 350 can
task its text processor 358 to send a text message (see arrow 3) to
the caller's device 10 confirming the issue to be resolved and/or
that a resolution will be forthcoming later via text message, in
step 237. The text message can include a ticket and/or tracking
number. The test message can be communicated to the caller device
10 via a multi-media messaging gateway 380.
[0062] In step 240, the deflection processor 352 tasks a
transcription processor 354 to transcribe the stored audio files
from the caller information, such as a caller voice message,
producing transcribed audio files and information. Transcribing the
spoken and/or recorded audio words can be done in real-time or from
stored audio files using, for example, a Transcription
Engine/Natural Language Engine, such as the Nuance Transcription
Engine (NTE) or other speech-to-text engine, to determine concepts
and intents associated with the call. The transcription processor
354 can assign a transcription confidence score to the
transcription.
[0063] In step 250, a classifier 355 analyzes the call information,
including the stored transcribed audio files, to determine concepts
and intents of the caller. The classifier 355 can implement various
techniques for analyzing and understanding concepts and intents of
the caller from at least the transcribed audio files. In some
embodiments, the classifier 355 can analyze the transcribed text
for the presence of meaningful and/or predetermined words, phrases,
and concepts in the transcription. In some embodiments, the
classifier 355 can be configured to utilize natural language
processing and understanding techniques and/or conversational
artificial intelligence techniques for determining the concepts and
intents.
[0064] The classifier 355 may further access account information of
an authenticated caller, which may be included or indicated in the
call information files, or accessible from backend systems 320.
Such account information could indicate an account status, a past
due account, a service interruption, a prior service call, a recent
change in service options, and so on. Such information can be used
as a meaningful context for interpreting the transcribed audio to
determine the concepts and intents and, ultimately, the issue to be
resolved.
[0065] The classifier 355 determines an appropriate skill group to
take responsibility for the call based on the determined contents
and intents. Once the skill group is determined, the deflection and
response system 350 can prepare and send information and files,
preferably grouped in an electronic folder, to a response system
360 including the selected skill group. The electronic folder can
include the transcribed text from the caller's audio, and one or
more of a transcription confidence score, the caller audio files, a
customer identification, authentication credentials, and other data
from the backend systems 320, as well as an identification of an
assigned skill group. The deflection and response system 350 can
prepare preformatted research templates (e.g., based on the
selected skill group) and a preformatted "canned" response template
with data-fill used to populate available response information.
These files can be included in the folder sent to the response
system 360. The assigned specialist from the selected skill group
will complete and finalize the response.
[0066] The response system 360 tasks a resolution processor 362 to
determine a resolution to the caller's issue, in step 260.
Depending on the caller's issue, as examples, the resolution can be
a solution to a problem, information satisfying an information
request, a service ordered, adjusted, or cancelled, a billing
adjustment, an account information update, instructional content,
or any other action or content responsive to the caller's
issue.
[0067] The resolution system 362 assigns responsibility for the
call to the skill group selected by the deflection and response
system 350. Within the selected skill group, a specialist (A, B, .
. . ) can take responsibility for the resolution. The resolution
system 362 can access backend systems 320 and knowledge bases 322
and/or third-party systems 370 to research the issue and determine
a resolution, candidate resolutions, and/or resolution options.
Such third-party systems 370 can include, but are not limited to,
information systems, diagnostic systems, subject matter expert
systems, and the like. The back end systems 320 and knowledge bases
322 and/or third-party systems 370 can serve as research
information sources, e.g., electrically accessible by the
resolution system 362.
[0068] In step 270, a text processor 364 generates and sends a text
message (arrow 4) to the caller's electronic device 10 with an
issue response communicating the resolution determined in step 260.
The test message can be communicated to the caller device 10 via a
multi-media messaging gateway 380.
[0069] In various embodiments, the response (arrow 4) to the caller
can be a SMS/MMS/RCS/iMessage text message or an Apple Business
Chat message that supports rich media, such as videos, GIFs, deep
links, and transmits in-call multimedia. The text message
preferably resolves the issue. But, in the event that the text
message does not provide a final resolution, the text message can
serve as a first message in a text or chat session between the
caller at the electronic device 10 and a chat entity at the
response system 360. Therefore, in some embodiments, the text
processor 364 can remain in a chat session with the electronic
device 10 until the session is terminated by the electronic device
10 or is timed out.
[0070] In some embodiments, the resolution system 362 and the text
processor 364 can include or take the form of a chat platform
including and/or accessible by one or more chat agents or chatbots,
as a chat entity. That is, in some embodiments, the response system
360 can be a chat platform 360. In some embodiments, the chat
platform 360 can take the form of or include, for example, the
Nina.RTM. platform provided by Nuance Communications, Inc., which
is an intelligent enterprise virtual assistant that integrates with
the popular internet-of-things (IoT).
[0071] The resolution system 362 can include a plurality of
different skill groups, each comprised of specialists (A, B, . .
.). The resolution system 362 can be configured to route the call
responsibility, and associated files, to a particular skill group
corresponding to the determined concepts and intents indicated by
the deflection and response system 350. For instance, a billing
issue can be routed to a billing skill group, a request for new
services can be routed to a sales skill/chat group, a technical
issue can be routed to a technician or engineer skill/chat group,
and so on. Within the appropriate skill/chat group, a particular
specialist A, B, . . . within the group, whether a chat agent or
chatbot, can take responsibility for resolution of the call.
[0072] Preferably, an assigned chat entity has access to and
training on all of the necessary billing, CRM systems, knowledge
bases, and other systems-of-record necessary to enable and empower
that chat entity to access enough information to resolve that
caller's issue in as few steps as possible, preferably with a first
and only text message (arrow 4). Such access can allow a chat agent
or a chatbot to personalize a text response based on that specific
and unique caller's intent and account. The result is an improved
caller experience and service delivered at a far lower cost. Each
successfully deflected call yields an immediate financial benefit
in the form of lower cost of transaction, and potentially a mid and
long-term benefit by moving the caller from a high-cost service
channel (Voice) to a lower cost digital channel on an ongoing
basis.
[0073] Therefore, provided is a deflection and response system and
method that deflect voice calls to an appropriate skill group and
specialist with information necessary to resolve a caller's issues
with, preferably, a single text message response. In some
embodiments, the caller can state a reason for the call and
responsibility for resolution of the issue raised by the caller can
be deflected to a response system 360 (e.g., a chat platform), for
text message response by a chat entity, e.g., a chat agent or a
chatbot. Therefore, according to the method, calls can be handled
by the IVR system 356 of the deflection and response system 350,
where the caller is prompted to provide a detailed reason for the
call, from which a context, concepts, caller intents, and/or caller
issue can be determined so that an assigned chat entity, e.g., a
chat agent/bot, can more quickly resolve the caller's issue.
[0074] In various embodiments, the caller's voice message is
transcribed by the transcription processor 354, e.g., using natural
language understanding (NLU) technology, to determine concepts and
intents of the call, from at least the transcribed audio. Based on
the determined concepts and intents, the deflection processor 352
routes responsibility for the call to the response system 360,
which includes a resolution system 362 and a text processor 364,
which can collectively form a chat platform. The chat platform can
include one or more chat skill groups, where each chat skill group
can have a predefined subject matter area of responsibility, skill,
or expertise. A chat skill group can be a specialized group that
could be predefined on the chat platform. A chat skill group can
include one or more chat entities, e.g., chat agents and/or
chatbots, which are available for resolving the issue. Chat
entities of a chat skill group can be specialists having
competency, skills, experience, and information access necessary to
resolve the caller's issue.
[0075] In various embodiments, the deflection and response system
350 and method 100/200 will: [0076] Transcribe the caller's voice
message [0077] Categorize intent/issue of the voice message using
an NLU classifier [0078] Optionally, authenticate the caller using
biometrics technology [0079] Obtain customer account information
and metadata from CRM, the caller's electronic device, or other
systems [0080] Assign the call, based on the intent/issue, to a
chat entity of a chat platform, such as a chatbot or chat agent
[0081] Send an immediate automated acknowledgement to the caller
device, advising the caller that its message was received, assigned
a chat entity, a "ticket/customer" number was assigned, and that a
reply to their issue will be sent when researched [0082] Research
the issue (e.g., question/request) [0083] Resolve caller's issue by
replying with a context-based text-message that includes a
personalized response, e.g., with free form answer and one-time
rich media link
[0084] In various embodiments, all of the above functions can be
automated, provided with minimal or no human interaction within the
deflection and response system 350.
[0085] Preferably, the chat platform 360 will have available to it
transcriptions, audio files, and customer information such that the
assigned chat entity will be able to research the caller's issue
and send a chat message to the caller, e.g., over SMS/RCS/iMessage,
with a response to the caller's issue. The objective is to provide
the chat entity with enough information to enable the chat entity
to resolve the caller's issue in one responsive text message.
[0086] In various embodiments, the transcription processor 354 can
be configured to generate the transcriptions with a transcription
confidence score. If the confidence score is above a certain
threshold, and an issue can be determined, the deflection and
response system 350 can text a message to the caller device 10 that
the issue is being worked. If the confidence score is below the
threshold, and an issue cannot be determined, the deflection and
response system 350 can communicate with the caller to obtain more
information or perform additional analysis.
[0087] The deflection and response system 350 can be configured to
deflect responsibility for the call based on the determined
concepts and intents to a particular skill group from a plurality
of available specialized skill groups. Each specialized skill group
can have a specific subject matter expertise, skill, and/or area of
responsibility. Each skill group can be a chat skill group on a
chat platform 360. A chat entity, e.g., a chatbot or chat agent,
within the specialized chat skill group can be associated with
resolving the caller's issue. In various embodiments, the chat
entity can receive at least the following information: [0088]
Asynchronous chat request [0089] Ticket/case/customer number [0090]
Transcribed text [0091] Issue Category/Intent/Summary [0092]
Customer info/container/metadata [0093] Audio files (e.g., in NTE
format) [0094] Transcription confidence score [0095] Preformatted
research template (e.g., based on issue category or subject matter)
[0096] Preformatted canned response with data-filled fields
[0097] A folder can be created that includes a set of files
associated with the call and the responsibility to resolve the
caller's issue. The set of files can include transcribed text from
audio, customer identification and authentication credentials, such
as CRM and other data, an .mp4 audio file, transcription confidence
score, and chat group or entity assigned. This folder can be routed
to a chat entity for resolution. One or more files, or the entire
folder, and can be encrypted.
[0098] In some embodiments, a separate process can be used create a
tracking ticket assigned for each transaction, i.e., caller issue.
The tracking ticket can be used by the customer if it needs to
follow up on the transaction. The tracking number and an
acknowledgement can be sent to the caller's electronic device 10
via an SMS/MMS/RCS/iMessage text message (or other form of text or
chat message) using a multi-media messaging gateway 380.
[0099] Once the issue is researched and a response determined, the
response system 360 (e.g., chat platform 360) sends the
personalized text message as a response (arrow 4) to the caller's
electronic device 10. In various embodiments, a chat session
remains (also arrow 4) with the chat entity or is reassigned and
transferred to a separate chat entity (or chat group) responsible
for keeping the chat open for the caller in case a further response
is immediately necessary. At this stage, the chat session can be
treated the same as other asynchronous chats on the chat platform
360.
[0100] FIG. 4 is a diagram depicting an embodiment of an audio
interaction between the caller and the deflection and response
system 350 of FIG. 3 using caller deice 10, in accordance with
aspects of the inventive concepts. The interaction is depicted as
arrow 2 in FIG. 3. In FIG. 4, an audio dialog is shown between the
caller on electronic device 10 and the IVR system 356 of the
deflection and response system 350. In 4-1, the IVR system 356
advises the caller of the wait time and offers the caller the
option to use the deflection and response system 350 (or the
"resolution and text message services"). In 4-2, the caller replies
"yes," indicating that the deflection and response system 350
should continue the dialog. In 4-3 and 4-4 the caller is
authenticated using an authentication processor, e.g.,
authentication processor 353. Authentication, such as voice
biometrics authentication, is optional and can be provided as
described herein above for example.
[0101] In 4-5, the caller is prompted by the IVR system 356 to
provide a complete description of the reasons for the call, which
is later transcribed to determine the issue needing resolution. In
4-6, the caller acknowledges the prompt and, in 4-7, the caller
states its message. In 4-8, the IVR system 356 acknowledges the
caller's audio message and provides a response indicating that the
issue resolution will be later texted to the caller's device
10.
[0102] FIGS. 5A through 5C depict an embodiment of text
interactions between the caller and the deflection and response
system 350 of FIG. 1, in accordance with aspects of the inventive
concepts. These figures depict an embodiment of a possible text
exchange between the deflection and response system 350 and the
caller's device 10 following the exchange between the IVR system
356 and the caller's device 10 in FIG. 4.
[0103] In FIG. 5A, the deflection and response system 350 sends an
acknowledgement text to the caller's device 10. The communication
is depicted by arrow 3 in FIG. 3. The acknowledgment text can be
sent by the response system/ chat platform 360. In this embodiment,
the text introduces a chat entity, here "Monica," and identifies
the perceived issue, in message 1/4. In message 2/4, the chat
platform 360 provides a ticket number. The issue was determined, at
least in part, based on the caller's voice message (see 4-7 in FIG.
4). This communication is depicted by arrow 3 in FIG. 3.
[0104] In FIG. 5B, in message 3/4, a new text from a specialized
chat entity (or skill group) of the chat platform 360, here "John,"
introduces himself and provides the problem resolution--second
movie charge is credited. As a chat entity, John can be a live chat
agent or a chat bot, as examples. This communication is depicted as
arrow 4 in FIG. 3.
[0105] In FIG. 5C, the chat platform 355 sends a follow-up message
4/4 indicating a further resolution of the issue--also waiving the
first movie charge. The second resolution message 4/4 is optional
and at the discretion of the chat entity. This communication is
also depicted as arrow 4 in FIG. 3.
[0106] Provided, therefore, are a call deflection and response
system and method; such systems and methods can deflect voice calls
to a specific skill group of a text response system, which is
configured to resolve caller's issue by replying with context-based
personalized response in the form of a SMS/MMS/RCS/iMessage text
message or chat message, as examples. Callers are allowed to leave
a detailed reason for their call with the IVR system of the
deflection and response system, concepts and intents are determined
from the caller's message. Based, at least in part, on the
determined concepts and intents, the call is directed to a selected
skill group of a chat platform to more quickly resolve the caller's
issue or request. The resolution is provided via text message to
the caller, e.g., to the caller's mobile phone.
[0107] Systems and methods in accordance with the inventive
concepts can be used broadly for all callers, callers on hold
beyond a certain time threshold, callers having a particular issue,
or they can be targeted to certain caller segments, such as high
value, high volume, millennials, or other defined caller
segments.
[0108] While the foregoing has described what are considered to be
the best mode and/or other preferred embodiments, it is understood
that various modifications can be made therein and that the
invention or inventions may be implemented in various forms and
embodiments, and that they may be applied in numerous applications,
only some of which have been described herein. It is intended by
the following claims to claim that which is literally described and
all equivalents thereto, including all modifications and variations
that fall within the scope of each claim.
[0109] It is appreciated that certain features of the invention,
which are, for clarity, described in the context of separate
embodiments, may also be provide in combination in a single
embodiment. Conversely, various features of the invention which
are, for brevity, described in the context of a single embodiment
may also be provided separately or in any suitable
sub-combination.
[0110] For example, it will be appreciated that all of the features
set out in any of the claims (whether independent or dependent) can
combined in any given way.
* * * * *