WO2020258168A1 - A method for managing incoming calls on a communication device and the corresponding device - Google Patents
A method for managing incoming calls on a communication device and the corresponding device Download PDFInfo
- Publication number
- WO2020258168A1 WO2020258168A1 PCT/CN2019/093305 CN2019093305W WO2020258168A1 WO 2020258168 A1 WO2020258168 A1 WO 2020258168A1 CN 2019093305 W CN2019093305 W CN 2019093305W WO 2020258168 A1 WO2020258168 A1 WO 2020258168A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- incoming call
- communication device
- rank
- caller
- conversational agent
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/436—Arrangements for screening incoming calls, i.e. evaluating the characteristics of a call before deciding whether to answer it
- H04M3/4365—Arrangements for screening incoming calls, i.e. evaluating the characteristics of a call before deciding whether to answer it based on information specified by the calling party, e.g. priority or subject
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/64—Automatic arrangements for answering calls; Automatic arrangements for recording messages for absent subscribers; Arrangements for recording conversations
- H04M1/642—Automatic arrangements for answering calls; Automatic arrangements for recording messages for absent subscribers; Arrangements for recording conversations storing speech in digital form
- H04M1/645—Automatic arrangements for answering calls; Automatic arrangements for recording messages for absent subscribers; Arrangements for recording conversations storing speech in digital form with speech synthesis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/527—Centralised call answering arrangements not requiring operator intervention
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/53—Centralised arrangements for recording incoming messages, i.e. mailbox systems
- H04M3/533—Voice mail systems
- H04M3/53308—Message originator indirectly connected to the message centre, e.g. after detection of busy or absent state of a called party
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2201/00—Electronic components, circuits, software, systems or apparatus used in telephone systems
- H04M2201/40—Electronic components, circuits, software, systems or apparatus used in telephone systems using speech recognition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2201/00—Electronic components, circuits, software, systems or apparatus used in telephone systems
- H04M2201/60—Medium conversion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2203/00—Aspects of automatic or semi-automatic exchanges
- H04M2203/65—Aspects of automatic or semi-automatic exchanges related to applications where calls are combined with other types of communication
- H04M2203/651—Text message transmission triggered by call
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/53—Centralised arrangements for recording incoming messages, i.e. mailbox systems
- H04M3/533—Voice mail systems
- H04M3/53333—Message receiving aspects
- H04M3/5335—Message type or catagory, e.g. priority, indication
Definitions
- the present invention relates to the management of incoming calls at a communication device. More particularly, the invention relates to voicemail management.
- a voicemail system is a computer-based system that allows users to exchange voice messages using communication devices such as telephones, mobile phones or smartphones.
- Voicemail systems are designed to convey recorded audio message from a caller to a callee. To do so, such voicemail systems comprise a user interface to select, play, and manage voice messages.
- the incoming call is forwarded toward the voicemail system of the communication device of the callee.
- the voicemail system can then answer the incoming call using the callee’s greeting message instructing the caller what comes next for example, "When you have finished recording, you may hang up or press hashtag for more options" .
- a method for managing an incoming call at a communication device comprising:
- the use of a conversational agent enables to analyze the speech of the caller and to gather relevant information from this speech.
- a rank is determined for the incoming calls. For example, incoming calls can be sorted into three category : nuisance, useful and important. In order to sort the incoming calls, keywords detected by the conversational agent in the speech of a caller are used.
- the conversational agent detects a keyword implying that the incoming call is an advertisement such as the word “reward” , then the incoming call is considered as a nuisance call.
- the conversational agent detects a keyword implying that the incoming call concerns a meeting such as the word “cinema” , then the incoming call is considered as a useful call.
- the conversational agent detects a keyword implying that the incoming call concerns the children of the callee such as the words “your child is ill” , then the incoming call is considered as an important call.
- actions are triggered on the communication device of the callee.
- the incoming call can be terminated.
- the incoming call can be passed to the callee so he/she can accept the call and directly speak with the caller.
- the conversational agent can be a chatbot capable of handling a conversation with the caller in order to gather information.
- a chatbot is capable of handling a conversation with the caller based on answers given by the caller to previous questions in order to get richer information.
- the solution according to the invention enables to provide the callee with more relevant and ore useful information regarding the incoming calls.
- the method further comprises :
- contextual information can be retrieved from sources other than the conversational agent.
- sources may be for example a contact list stored in the communication device of the callee which comprises identifiers associated with communication devices of callers known from the callee, a black list comprising identifiers associated with callers the callee does not want to be contacted by, or by consulting an internet website found using information gathered by the conversational agent.
- obtaining information from the caller comprises detecting the voiceprint of the caller.
- Voiceprints of potential callers may be stored in the callee device, in order to recognize the caller’s identity. Such a solution enables to trigger action specifically associated with the caller as function of their identity.
- the caller may be handled in a custom way by the conversational agent which contributes to the user-friendliness of the invention.
- a list of at least one action to be triggered is associated with each rank, and the method further comprises selecting, in the list of actions associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device.
- the selection, in the list associated with the determined rank of said incoming call, of said at least one action to be triggered on said communication device is based at least on said information obtained by the conversational agent
- At least one or more actions can be triggered on the communication device of the callee. For example, when an incoming call is ranked as useful and that the conversational agent determines that such incoming call concerns a meeting for watching a movie, the date and time of the movie can be stored in a calendar application of the communication device and a voice message containing all the information related to this meeting can be stored in the voice message system of the communication device of the callee so the callee can listen to this voice message.
- the actions belong to a group comprising :
- the invention also concerns a communication device capable of managing an incoming call, said communication device comprising :
- a list of at least one action to be triggered being associated with each rank, and the at least one processor is further configured to select, in the list associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device.
- the at least one processor is further configured to select, in the list associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device based at least on said information obtained by the conversational agent.
- Some processes implemented by elements of the invention may be computer implemented. Accordingly, such elements may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc. ) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit” , "module” or “system' . Furthermore, such elements may take the form of a computer program product embodied in any tangible medium of expression having computer usable program code embodied in the medium.
- a tangible carrier medium may comprise a storage medium such as a floppy disk, a CD-ROM, a hard disk drive, a magnetic tape device or a solid-state memory device and the like.
- a transient carrier medium may include a signal such as an electrical signal, an electronic signal, an optical signal, an acoustic signal, a magnetic signal or an electromagnetic signal, e.g. a microwave or RF signal.
- Figure 1 represents a flowchart of the steps of the method for managing an incoming call when the method is executed by a communication device called callee device;
- Figure 2 represents a detailed view of one of the callee device according to an embodiment of the invention.
- Figure 1 represents a flowchart of the steps of the method for managing an incoming call when the method is executed by a communication device, here called “callee device CeeD” .
- a communication device here called “callee device CeeD” .
- Such a callee device CeeD is for example a smartphone.
- the method for managing an incoming call according to an embodiment of the invention can be implement as a part of a modified voicemail module that can be downloaded and installed on the callee device CeeD when software implemented.
- the callee device CeeD receives a request for establishing a communication session from another communication device called “caller device CerD” .
- the callee device CeeD transmits to the caller device CerD an acknowledgment message completing the establishment of the communication session between the caller device CerD and the callee device CeeD.
- an incoming call is placed by the caller device CerD toward the callee device CeeD.
- the voicemail module associated with the callee device CeeD is activated in a step E3 together with a conversation agent embedded in the callee device CeeD.
- the conversational agent analyses a speech of a user of the caller device CerD during the incoming call, in order to obtain relevant information.
- the conversational agent can listen to the speech of the user of the caller device CerD in order to detect keywords or it can engage the user of the caller device CerD in a conversation based on keywords detected in the answers provided by the user of the caller device CerD.
- the voice recognition and speech language understanding sub-steps Thanks to the voice recognition and speech language understanding sub-steps, a transcription of what the caller says during the incoming call can be accurately obtained. And thanks to the dialogue management sub-step, the “intent” of the caller can be derived from the transcription, as information to be used later in the present method.
- a first implementation of the dialogue management sub-step relies on a “key Phrase &Keyword” approach, wherein keyword (s) or a key phrase associated with specific caller intent is detected in the text, then the associated caller intent is deduced from these detected keyword (s) or key phrase.
- dialog management sub-step relies on using a state machine diagram of dialog flow, wherein the conversational agent maintains a state diagram: when the conversational agent detects what the caller is saying, it will use the corresponding sentence in this state diagram as an answer.
- the dialog could be as follow:
- the caller doesn’ t answer the first question “Who are you? ” but transfers to another topic “What will you do? ” .
- the conversational agent detects this topic transfer and adapts its next question accordingly in order to obtain some additional information about/from the caller.
- the conversational agent can detect the voiceprint of the caller when the caller speaks during the incoming call.
- voiceprints of potential callers may be stored in the callee device CeeD, in order to compare the voiceprint of the caller and recognize the caller’s identity from this comparison, and therefore obtain information about the caller’s identity.
- action specifically associated with the caller can be triggered directly.
- the conversational agent can detect that the callee’s boss is calling, thanks to the voiceprint recognition. And this information about the caller’s identity can then be used to deduce that the incoming call is important.
- the conversational agent may rely on machine learning, in other words the more it converses with users of distinct caller devices CerD, the more it learns.
- the conversational agent can be trained on specific datasets enabling it to sort the incoming calls.
- federated learning can be used by the conversational agent, which is an approach to machine learning where a shared global model is trained across many participating clients that keep their training data locally.
- TDF TensorFlow Federated
- This enables to improve the quality of the analysis performed by the conversational agent and consequently the relevance of the decisions taken based on the collected information. What’s more, when using this technology, sensitive data (for example the speech data or transcription data) is not uploaded to remote servers.
- contextual information can be retrieved by the voicemail module from sources other than the conversational agent.
- sources may be for example a contact list stored in the callee device CeeD.
- a contact list comprises identifiers associated with communication devices of callers known from the callee.
- the source may be a black list comprising identifiers associated with callers the callee does not want to be contacted by.
- Another example of such a source can be an internet website found using information gathered by the conversational agent.
- the voicemail module determines a rank for said incoming call based on the information obtained by the conversation agent during step E4 and possibly the contextual information collected during step E6.
- a set of three ranks can be defined : “nuisance” , “useful” and “important” .
- keywords detected by the conversational agent in the caller’s speech during the incoming call are used :
- the conversational agent detects a keyword implying that the incoming call is an advertisement, such as the word “reward” , then the incoming call is considered as a spam and is associated with the rank “nuisance” ,
- the conversational agent detects a keyword implying that the incoming call concerns a meeting, such as the word “cinema” , then the incoming call is considered relevant and is associated with the rank “useful” , or
- this incoming call is considered as an important call and is associated with the rank “important” .
- a set of five ranks can be defined.
- a call concerning an advertisement can be associated with a high rank, i.e. a rank of 5, a call concerning a date at a nearby theatre can be ranked 3, a call concerning an appointment with a manager of the callee can be ranked 2 while an urgent call, e.g. a call concerning an ill child, can be associated with a rank of 1.
- the voicemail module selects at least one action regarding to the incoming call to be triggered based on the rank of the incoming call determined during step E6 and based on the information collected by the conversational agent and on the contextual information.
- the actions to be triggered on the callee device CeeD are stored in one or more lists of actions stored in a memory of the callee device CeeD. There can be as many lists of actions as there are ranks of incoming calls. A same action can be stored in one or more lists of actions. Such actions can be for example:
- the user of the callee device CeeD may configure his/her own custom actions based on his/her needs and store these custom actions in the callee device CeeD, in the different lists of actions that can be triggered.
- the voicemail module selects, amongst the plurality of available lists of actions, the list of actions in which the actions be triggered can be selected. Indeed, an incoming call associated with the rank “nuisance” is not handled the same way as an incoming call ranked “important” or ranked 1.
- the voicemail module selects one or more actions to be triggered based on the information collected by the conversational agent and possibly the contextual information.
- an action can be to ask the caller to leave a voice message or to express “I’ m not interested in it “and end the call, without forwarding the incoming call to the callee.
- the selected actions can be :
- an action can be for the conversational agent to ask the caller to wait as the voicemail module tries to get the callee to answer the incoming call. If the callee answers the call, a dialog box with some summary of the information collected by the conversational agent and possibly the contextual information can be displayed on a screen of the callee device CeeD, which helps the callee understands the context of the incoming call.
- step E8 the selected action (s) is triggered.
- Figure 2 represents a detailed view of one of the callee device CeeD according to an embodiment of the invention.
- a callee device CeeD may comprise at least one hardware processor 201, a storage unit 202, an input device 203, a display device 204, an interface unit 205, a network interface 206, and a voicemail module 207 which are typically connected by a data bus 208.
- constituent elements of the callee device CeeD may be connected by a connection other than a data bus connection.
- the hardware processor (s) 201 controls operations of the callee device CeeD.
- the storage unit 202 stores at least one program capable of managing an incoming call to be executed by the processor 201, and various data, such as parameters used by computations performed by the processor 201, intermediate data of computations performed by the processor 201, and so on.
- the processor 201 may be formed by any known and suitable hardware, or software, or a combination of hardware and software.
- the processor 201 may be formed by dedicated hardware such as a processing circuit, or by a programmable processing unit such as a CPU (Central Processing Unit) that executes a program stored in a memory thereof.
- CPU Central Processing Unit
- the storage unit 202 may be formed by any suitable storage or means capable of storing the program, data, or the like in a computer-readable manner. Examples of the storage unit 202 include non-transitory computer-readable storage media such as semiconductor memory devices, and magnetic, optical, or magneto-optical recording media loaded into a read and write unit.
- the program causes the processor 201 to perform a process according to an embodiment of the present invention as described with reference to figure 1.
- the input device 203 may be formed by a keyboard, a pointing device such as a mouse, or the like for use by the user to input commands, for example to make user's selections of parameters used for selecting the transmission interface to be used.
- the display device 204 may be formed by a display device to display, for example, a Graphical User Interface (GUI) .
- GUI Graphical User Interface
- the input device 203 and the display device 204 may be formed integrally by a touchscreen panel, for example.
- the interface unit 205 provides an interface between the callee device CeeD and an external apparatus.
- the interface unit 205 may be communicable with the external apparatus via cable or wireless communication.
- Such an external apparatus is for example a home gateway.
- a network interface 206 provides a connection between the callee device CeeD and the caller device CerD via a backbone network (not shown in the figures) , such as the Internet.
- the network interface 206 may provide, depending on its nature, a wired or a wireless connection to the backbone network.
- the voicemail module 207 comprises a conversational agent, which is capable of analyzing speeches from users of caller devices CerD during an incoming call toward the callee device CeeD, of ranking the incoming calls toward the callee device CeeD and triggering different actions on the callee device CeeD depending on the speech analysis, as explained previously.
- a voicemail module 207 is typically implemented as a computer program which may be stored in storage unit 202 and comprises code instructions which can be executed by the hardware processor (s) 201 in order to perform this speech analysis, incoming calls ranking and action triggering operations.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Telephone Function (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention relates to a method for managing an incoming call at a communication device (CeeD). In such a method, information from a caller whose communication device is causing said incoming call is obtained (E4) by means of a conversational agent, a rank is determined (E6) for said incoming call based on at least said information obtained by the conversational agent, and at least one action is triggered (E7), on said communication device, regarding said incoming call based on the rank of the incoming call. The invention also relates to a communication device for carrying out such a method, as well as to a corresponding computer program.
Description
The present invention relates to the management of incoming calls at a communication device. More particularly, the invention relates to voicemail management.
A voicemail system is a computer-based system that allows users to exchange voice messages using communication devices such as telephones, mobile phones or smartphones.
Voicemail systems are designed to convey recorded audio message from a caller to a callee. To do so, such voicemail systems comprise a user interface to select, play, and manage voice messages.
As it is well known, when an incoming call comes in on a communication device but the user of said communication device, here after the callee, doesn't answer, the incoming call is forwarded toward the voicemail system of the communication device of the callee. The voicemail system can then answer the incoming call using the callee’s greeting message instructing the caller what comes next for example, "When you have finished recording, you may hang up or press hashtag for more options" .
Such voicemail systems, however, present some drawbacks. The caller tends not to leave messages or when they do so, the messages are, more often than not, short and concise. As a consequence, the information provided to the callee may lack context and/or may not be enough for the callee to gather enough information from the message.
It would hence be desirable to provide a solution enabling more information to be gathered from the caller when the incoming call is forwarded toward the voicemail of the callee’s communication device, thus providing the callee with more relevant information regarding the incoming calls while keeping the interaction between the caller and the voicemail system user-friendly.
3. Summary
According to a first aspect of the invention, there is provided a method for managing an incoming call at a communication device comprising :
- obtaining, by means of a conversational agent, information from a caller whose communication device is causing said incoming call,
- determining a rank for said incoming call based on at least said information obtained by the conversational agent,
- triggering at least one action, on said communication device, regarding said incoming call based on the rank of the incoming call.
The use of a conversational agent enables to analyze the speech of the caller and to gather relevant information from this speech.
Based on the gathered information, a rank is determined for the incoming calls. For example, incoming calls can be sorted into three category : nuisance, useful and important. In order to sort the incoming calls, keywords detected by the conversational agent in the speech of a caller are used.
If the conversational agent detects a keyword implying that the incoming call is an advertisement such as the word “reward” , then the incoming call is considered as a nuisance call.
If the conversational agent detects a keyword implying that the incoming call concerns a meeting such as the word “cinema” , then the incoming call is considered as a useful call.
If the conversational agent detects a keyword implying that the incoming call concerns the children of the callee such as the words “your child is ill” , then the incoming call is considered as an important call.
Depending on the rank associated with the incoming call, actions are triggered on the communication device of the callee. Thus, when the incoming call is a nuisance call, the incoming call can be terminated. When the incoming call is an important call, the incoming call can be passed to the callee so he/she can accept the call and directly speak with the caller.
Other actions may be triggered on the communication device depending on the rank associated with the incoming call.
The conversational agent can be a chatbot capable of handling a conversation with the caller in order to gather information. Such a chatbot is capable of handling a conversation with the caller based on answers given by the caller to previous questions in order to get richer information.
By sorting the incoming calls based on information gathered by means of a conversation agent and by triggering actions on the communication device of the callee based on the ranks associated with the incoming calls, the solution according to the invention enables to provide the callee with more relevant and ore useful information regarding the incoming calls.
In an embodiment of the invention, the method further comprises :
- obtaining contextual information related to said incoming call;wherein the rank of the incoming call is determined based on said information obtained by the conversational agent and on said contextual information.
In order to determine a rank to associated with an incoming call more accurately, contextual information can be retrieved from sources other than the conversational agent. Such sources may be for example a contact list stored in the communication device of the callee which comprises identifiers associated with communication devices of callers known from the callee, a black list comprising identifiers associated with callers the callee does not want to be contacted by, or by consulting an internet website found using information gathered by the conversational agent.
According to a feature of the invention, obtaining information from the caller comprises detecting the voiceprint of the caller.
Voiceprints of potential callers may be stored in the callee device, in order to recognize the caller’s identity. Such a solution enables to trigger action specifically associated with the caller as function of their identity. The caller may be handled in a custom way by the conversational agent which contributes to the user-friendliness of the invention.
In yet another embodiment of the invention, a list of at least one action to be triggered is associated with each rank, and the method further comprises selecting, in the list of actions associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device.
The selection, in the list associated with the determined rank of said incoming call, of said at least one action to be triggered on said communication device is based at least on said information obtained by the conversational agent
Depending on the rank associated with an incoming call, at least one or more actions can be triggered on the communication device of the callee. For example, when an incoming call is ranked as useful and that the conversational agent determines that such incoming call concerns a meeting for watching a movie, the date and time of the movie can be stored in a calendar application of the communication device and a voice message containing all the information related to this meeting can be stored in the voice message system of the communication device of the callee so the callee can listen to this voice message.
According to yet another aspect of the invention, the actions belong to a group comprising :
- terminating said incoming call,
- recording a conversation engaged by said conversational agent with the caller,
- converting information collected by said conversational agent into a text message to be presented to a user of said communication device,
- recording an event in an agenda application loaded in said communication device.
The invention also concerns a communication device capable of managing an incoming call, said communication device comprising :
- a conversational agent capable of obtaining information from a caller whose communication device is causing said incoming call,
and at least one processor configured to :
- determine a rank for said incoming call based on at least said information obtained by the conversational agent,
- trigger at least one action, on said communication device, regarding said incoming call based on the rank of the incoming call.
In an embodiment of the communication device according to the invention a list of at least one action to be triggered being associated with each rank, and the at least one processor is further configured to select, in the list associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device.
In another embodiment of the communication device, the at least one processor is further configured to select, in the list associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device based at least on said information obtained by the conversational agent.
Some processes implemented by elements of the invention may be computer implemented. Accordingly, such elements may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc. ) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit" , "module" or "system' . Furthermore, such elements may take the form of a computer program product embodied in any tangible medium of expression having computer usable program code embodied in the medium.
Since elements of the present invention can be implemented in software, the present invention can be embodied as computer readable code for provision to a programmable apparatus on any suitable carrier medium. A tangible carrier medium may comprise a storage medium such as a floppy disk, a CD-ROM, a hard disk drive, a magnetic tape device or a solid-state memory device and the like. A transient carrier medium may include a signal such as an electrical signal, an electronic signal, an optical signal, an acoustic signal, a magnetic signal or an electromagnetic signal, e.g. a microwave or RF signal.
The present invention can be better understood with reference to the following description and drawings, given by way of example and not limiting the scope of protection, and in which:
Figure 1 represents a flowchart of the steps of the method for managing an incoming call when the method is executed by a communication device called callee device; and
Figure 2 represents a detailed view of one of the callee device according to an embodiment of the invention.
The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention.
The present description illustrates the principles of the present invention. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described or shown herein, embody the principles of the invention and are included within its spirit and scope.
Figure 1 represents a flowchart of the steps of the method for managing an incoming call when the method is executed by a communication device, here called “callee device CeeD” . Such a callee device CeeD is for example a smartphone. The method for managing an incoming call according to an embodiment of the invention can be implement as a part of a modified voicemail module that can be downloaded and installed on the callee device CeeD when software implemented.
In a step E1, the callee device CeeD receives a request for establishing a communication session from another communication device called “caller device CerD” .
In a step E2, the callee device CeeD transmits to the caller device CerD an acknowledgment message completing the establishment of the communication session between the caller device CerD and the callee device CeeD. Once the communication session is established between the caller device CerD and the callee device CeeD, an incoming call is placed by the caller device CerD toward the callee device CeeD.
In the meantime, the voicemail module associated with the callee device CeeD is activated in a step E3 together with a conversation agent embedded in the callee device CeeD.
In a step E4, the conversational agent analyses a speech of a user of the caller device CerD during the incoming call, in order to obtain relevant information. Depending on the implementation, the conversational agent can listen to the speech of the user of the caller device CerD in order to detect keywords or it can engage the user of the caller device CerD in a conversation based on keywords detected in the answers provided by the user of the caller device CerD.
In order to carry out this speech analysis step E4, conventional speech analysis technologies can be used, which may separate this keyword detection process into three sub-steps: voice recognition, speech language understanding and dialogue management.
Thanks to the voice recognition and speech language understanding sub-steps, a transcription of what the caller says during the incoming call can be accurately obtained. And thanks to the dialogue management sub-step, the “intent” of the caller can be derived from the transcription, as information to be used later in the present method.
A first implementation of the dialogue management sub-step relies on a “key Phrase &Keyword” approach, wherein keyword (s) or a key phrase associated with specific caller intent is detected in the text, then the associated caller intent is deduced from these detected keyword (s) or key phrase.
Another implementation of the dialogue management sub-step relies on using a state machine diagram of dialog flow, wherein the conversational agent maintains a state diagram: when the conversational agent detects what the caller is saying, it will use the corresponding sentence in this state diagram as an answer. For example, the dialog could be as follow:
“Conversational agent: Hello, I’ m in the meeting now, Who are you?
Caller: I’ d like to go to Cinema with you this Tuesday morning.
Conversational agent: Go to Cinema on Tuesday. What is your full name?
Caller: Paul. ”
In this example, the caller doesn’ t answer the first question “Who are you? ” but transfers to another topic “What will you do? ” . The conversational agent detects this topic transfer and adapts its next question accordingly in order to obtain some additional information about/from the caller.
In another embodiment, which can be combined with the above-mentioned keyword/key phrase detection or the conversation embodiments, the conversational agent can detect the voiceprint of the caller when the caller speaks during the incoming call. In this embodiment, voiceprints of potential callers may be stored in the callee device CeeD, in order to compare the voiceprint of the caller and recognize the caller’s identity from this comparison, and therefore obtain information about the caller’s identity. Upon recognition of the caller identity, action specifically associated with the caller can be triggered directly.
For example, when a person is called by his/her boss during a business trip, the conversational agent can detect that the callee’s boss is calling, thanks to the voiceprint recognition. And this information about the caller’s identity can then be used to deduce that the incoming call is important. In an embodiment of the invention, the conversational agent may rely on machine learning, in other words the more it converses with users of distinct caller devices CerD, the more it learns. The conversational agent can be trained on specific datasets enabling it to sort the incoming calls.
In particular, a so-called “federated learning” can be used by the conversational agent, which is an approach to machine learning where a shared global model is trained across many participating clients that keep their training data locally. For instance, the TensorFlow Federated (TFF) open-source framework can be used here. This enables to improve the quality of the analysis performed by the conversational agent and consequently the relevance of the decisions taken based on the collected information. What’s more, when using this technology, sensitive data (for example the speech data or transcription data) is not uploaded to remote servers.
In a step E5, which is optional, contextual information can be retrieved by the voicemail module from sources other than the conversational agent. Such sources may be for example a contact list stored in the callee device CeeD. Such a contact list comprises identifiers associated with communication devices of callers known from the callee. In another embodiment, the source may be a black list comprising identifiers associated with callers the callee does not want to be contacted by.
Another example of such a source can be an internet website found using information gathered by the conversational agent.
In a step E6, the voicemail module determines a rank for said incoming call based on the information obtained by the conversation agent during step E4 and possibly the contextual information collected during step E6.
In a first example a set of three ranks can be defined : “nuisance” , “useful” and “important” . In order to associate the incoming calls with one of these three ranks, keywords detected by the conversational agent in the caller’s speech during the incoming call are used :
- when the conversational agent detects a keyword implying that the incoming call is an advertisement, such as the word “reward” , then the incoming call is considered as a spam and is associated with the rank “nuisance” ,
- when the conversational agent detects a keyword implying that the incoming call concerns a meeting, such as the word “cinema” , then the incoming call is considered relevant and is associated with the rank “useful” , or
- when the conversational agent detects keywords implying that the incoming call concerns the children of the callee, such as the words “your child is ill” , or when the voiceprint of the callee’s boss is detected in the incoming call, then this incoming call is considered as an important call and is associated with the rank “important” .
In a second example, a set of five ranks can be defined. For example, a call concerning an advertisement can be associated with a high rank, i.e. a rank of 5, a call concerning a date at a nearby theatre can be ranked 3, a call concerning an appointment with a manager of the callee can be ranked 2 while an urgent call, e.g. a call concerning an ill child, can be associated with a rank of 1.
In a step E7, the voicemail module selects at least one action regarding to the incoming call to be triggered based on the rank of the incoming call determined during step E6 and based on the information collected by the conversational agent and on the contextual information.
The actions to be triggered on the callee device CeeD are stored in one or more lists of actions stored in a memory of the callee device CeeD. There can be as many lists of actions as there are ranks of incoming calls. A same action can be stored in one or more lists of actions. Such actions can be for example:
- terminating said incoming call,
- recording a conversation engaged by said conversational agent with the caller,
- converting information collected by said conversational agent into a text message to be presented to a user of said communication device,
- recording an event in an agenda application loaded in said communication device.
The user of the callee device CeeD may configure his/her own custom actions based on his/her needs and store these custom actions in the callee device CeeD, in the different lists of actions that can be triggered.
Thus, based on the rank associated with the incoming call during the step E6, the voicemail module selects, amongst the plurality of available lists of actions, the list of actions in which the actions be triggered can be selected. Indeed, an incoming call associated with the rank “nuisance” is not handled the same way as an incoming call ranked “important” or ranked 1.
Once the list of actions to be triggered on the callee device CeeD is selected, the voicemail module selects one or more actions to be triggered based on the information collected by the conversational agent and possibly the contextual information.
For instance, in the case of an incoming call ranked “nuisance” , an action can be to ask the caller to leave a voice message or to express “I’ m not interested in it “and end the call, without forwarding the incoming call to the callee.
When an incoming call is ranked “useful” and that the conversational agent determines that such incoming call concerns a meeting for watching a movie, the selected actions can be :
- store in a calendar application of the communication device the date and time of the movie ,
- generate a voice message containing all the information related to this meeting and store the generated message in the voice message system of the communication device of the callee so the callee can listen to this voice message,
- trigger a localization application in which the location of the cinema is pinpointed.
When the incoming call is ranked ” important” , an action can be for the conversational agent to ask the caller to wait as the voicemail module tries to get the callee to answer the incoming call. If the callee answers the call, a dialog box with some summary of the information collected by the conversational agent and possibly the contextual information can be displayed on a screen of the callee device CeeD, which helps the callee understands the context of the incoming call.
In a step E8, the selected action (s) is triggered.
Figure 2 represents a detailed view of one of the callee device CeeD according to an embodiment of the invention.
A callee device CeeD may comprise at least one hardware processor 201, a storage unit 202, an input device 203, a display device 204, an interface unit 205, a network interface 206, and a voicemail module 207 which are typically connected by a data bus 208. Of course, constituent elements of the callee device CeeD may be connected by a connection other than a data bus connection.
The hardware processor (s) 201 controls operations of the callee device CeeD. The storage unit 202 stores at least one program capable of managing an incoming call to be executed by the processor 201, and various data, such as parameters used by computations performed by the processor 201, intermediate data of computations performed by the processor 201, and so on. The processor 201 may be formed by any known and suitable hardware, or software, or a combination of hardware and software. For example, the processor 201 may be formed by dedicated hardware such as a processing circuit, or by a programmable processing unit such as a CPU (Central Processing Unit) that executes a program stored in a memory thereof.
The storage unit 202 may be formed by any suitable storage or means capable of storing the program, data, or the like in a computer-readable manner. Examples of the storage unit 202 include non-transitory computer-readable storage media such as semiconductor memory devices, and magnetic, optical, or magneto-optical recording media loaded into a read and write unit. The program causes the processor 201 to perform a process according to an embodiment of the present invention as described with reference to figure 1.
The input device 203 may be formed by a keyboard, a pointing device such as a mouse, or the like for use by the user to input commands, for example to make user's selections of parameters used for selecting the transmission interface to be used. The display device 204 may be formed by a display device to display, for example, a Graphical User Interface (GUI) . The input device 203 and the display device 204 may be formed integrally by a touchscreen panel, for example.
The interface unit 205 provides an interface between the callee device CeeD and an external apparatus. The interface unit 205 may be communicable with the external apparatus via cable or wireless communication. Such an external apparatus is for example a home gateway.
A network interface 206 provides a connection between the callee device CeeD and the caller device CerD via a backbone network (not shown in the figures) , such as the Internet. The network interface 206 may provide, depending on its nature, a wired or a wireless connection to the backbone network.
The voicemail module 207 comprises a conversational agent, which is capable of analyzing speeches from users of caller devices CerD during an incoming call toward the callee device CeeD, of ranking the incoming calls toward the callee device CeeD and triggering different actions on the callee device CeeD depending on the speech analysis, as explained previously. Such a voicemail module 207 is typically implemented as a computer program which may be stored in storage unit 202 and comprises code instructions which can be executed by the hardware processor (s) 201 in order to perform this speech analysis, incoming calls ranking and action triggering operations.
Although the present invention has been described hereinabove with reference to specific embodiments, the present invention is not limited to the specific embodiments, and modifications will be apparent to a skilled person in the art which lie within the scope of the present invention.
Many further modifications and variations will suggest themselves to those versed in the art upon making reference to the foregoing illustrative embodiments, which are given by way of example only and which are not intended to limit the scope of the invention, that being determined solely by the appended claims. In particular the different features from different embodiments may be interchanged, where appropriate.
Claims (11)
- A method for managing an incoming call at a communication device (CeeD) comprising :- obtaining (E4) , by means of a conversational agent, information from a caller whose communication device is causing said incoming call,- determining (E6) a rank for said incoming call based on at least said information obtained by the conversational agent, and- triggering (E7) at least one action, on said communication device, regarding said incoming call based on the rank of the incoming call.
- The method for managing an incoming call according to claim 1, further comprising :- obtaining (E5) contextual information related to said incoming call;wherein the rank of the incoming call is determined based on said information obtained by the conversational agent and on said contextual information.
- The method for managing an incoming call according to claim 1 or 2, in which obtaining information from the caller comprises the conversational agent engaging the caller in a conversation.
- The method for managing an incoming call according to any one of claims 1 to 3, in which obtaining information from the caller comprises detecting the voiceprint of the caller.
- The method for managing an incoming call according to any one of claims 1 to 4, wherein a list of at least one action to be triggered is associated with each rank, the method further comprising selecting, in the list of actions associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device.
- The method for managing an incoming call according to claim 5, wherein the selection, in the list associated with the determined rank of said incoming call, of said at least one action to be triggered on said communication device is based at least on said information obtained by the conversational agent.
- The method for managing an incoming call according to any one of claims 1 to 6 wherein the actions belong to a group comprising :- terminating said incoming call,- recording a conversation engaged by said conversational agent with the caller,- converting information collected by said conversational agent into a text message to be presented to a user of said communication device,- recording an event in an agenda application loaded in said communication device.
- A communication device capable of managing an incoming call, said communication device comprising a conversational agent, capable of obtaining information from a caller whose communication device is causing said incoming call, and at least one processor configured to :- determine a rank for said incoming call based on at least said information obtained by the conversational agent,- trigger at least one action, on said communication device, regarding said incoming call based on the rank of the incoming call.
- The communication device according to claim 8 wherein, a list of at least one action to be triggered being associated with each rank, the at least one processor is further configured to select, in the list associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device.
- The communication device according to claim 9, wherein the at least one processor is further configured to select, in the list associated with the determined rank of said incoming call, said at least one action to be triggered on said communication device based at least on said information obtained by the conversational agent.
- A computer program characterized in that it comprises program code instructions for the implementation of the method according to any one of claims 1 to 7, when the computer program is executed by a processor.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2019/093305 WO2020258168A1 (en) | 2019-06-27 | 2019-06-27 | A method for managing incoming calls on a communication device and the corresponding device |
PCT/IB2020/000585 WO2020260954A1 (en) | 2019-06-27 | 2020-06-24 | A method for managing incoming calls on a communication device and the corresponding device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2019/093305 WO2020258168A1 (en) | 2019-06-27 | 2019-06-27 | A method for managing incoming calls on a communication device and the corresponding device |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020258168A1 true WO2020258168A1 (en) | 2020-12-30 |
Family
ID=72039618
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/093305 WO2020258168A1 (en) | 2019-06-27 | 2019-06-27 | A method for managing incoming calls on a communication device and the corresponding device |
PCT/IB2020/000585 WO2020260954A1 (en) | 2019-06-27 | 2020-06-24 | A method for managing incoming calls on a communication device and the corresponding device |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/IB2020/000585 WO2020260954A1 (en) | 2019-06-27 | 2020-06-24 | A method for managing incoming calls on a communication device and the corresponding device |
Country Status (1)
Country | Link |
---|---|
WO (2) | WO2020258168A1 (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100142684A1 (en) * | 2008-12-04 | 2010-06-10 | At&T Intellectual Property I, L.P. | System and method for automatically transcribing voicemail |
US20150003334A1 (en) * | 2013-06-26 | 2015-01-01 | Sap Ag | Prioritized message notification for mobile communication devices |
US20190020762A1 (en) * | 2017-07-13 | 2019-01-17 | Lenovo (Singapore) Pte. Ltd. | Systems and methods to parse message for providing alert at device |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018013804A1 (en) * | 2016-07-15 | 2018-01-18 | Circle River, Inc. | Automated call answering based on artificial intelligence |
US10171675B1 (en) * | 2016-10-10 | 2019-01-01 | Open Invention Network, Llc | Automated intelligent personal representative |
KR102479499B1 (en) * | 2017-11-22 | 2022-12-21 | 엘지전자 주식회사 | Mobile terminal |
-
2019
- 2019-06-27 WO PCT/CN2019/093305 patent/WO2020258168A1/en active Application Filing
-
2020
- 2020-06-24 WO PCT/IB2020/000585 patent/WO2020260954A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100142684A1 (en) * | 2008-12-04 | 2010-06-10 | At&T Intellectual Property I, L.P. | System and method for automatically transcribing voicemail |
US20150003334A1 (en) * | 2013-06-26 | 2015-01-01 | Sap Ag | Prioritized message notification for mobile communication devices |
US20190020762A1 (en) * | 2017-07-13 | 2019-01-17 | Lenovo (Singapore) Pte. Ltd. | Systems and methods to parse message for providing alert at device |
Also Published As
Publication number | Publication date |
---|---|
WO2020260954A1 (en) | 2020-12-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11349991B2 (en) | Systems and methods to present voice message information to a user of a computing device | |
US8311188B2 (en) | User interface with voice message summary | |
CN112470454A (en) | Synchronous communication using voice and text | |
US20210134284A1 (en) | Methods and systems for personalized virtual agents to learn from customers | |
US20210134283A1 (en) | Methods and systems of virtual agent real-time recommendation, suggestion and advertisement | |
US20210133765A1 (en) | Methods and systems for socially aware virtual agents | |
JP7297797B2 (en) | Method and apparatus for managing holds | |
US20220334795A1 (en) | System and method for providing a response to a user query using a visual assistant | |
US10257350B2 (en) | Playing back portions of a recorded conversation based on keywords | |
US20210136195A1 (en) | Methods and systems for virtual agent to understand and detect spammers, fraud calls, and auto dialers | |
US20210136208A1 (en) | Methods and systems for virtual agent to understand and detect spammers, fraud calls, and auto dialers | |
US20210136209A1 (en) | Methods and systems for virtual agents to check caller identity via multi channels | |
KR20240046433A (en) | Determine and display estimated hold period for calls | |
US9042523B2 (en) | Modifying voice messages stored in a voicemail system | |
WO2020258168A1 (en) | A method for managing incoming calls on a communication device and the corresponding device | |
JPWO2012124116A1 (en) | Call management device, call management method, and program | |
EP2680256A1 (en) | System and method to analyze voice communications | |
US20210133764A1 (en) | Methods and systems for virtual agents to check caller identity via multi channels |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19935657 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19935657 Country of ref document: EP Kind code of ref document: A1 |