CN109964473B

CN109964473B - Voice service response method and device

Info

Publication number: CN109964473B
Application number: CN201780058753.XA
Authority: CN
Inventors: 吴震
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2017-04-01
Filing date: 2017-07-28
Publication date: 2021-03-05
Anticipated expiration: 2037-07-28
Also published as: CN109964473A; WO2018176705A1

Abstract

The application provides a voice service response method and a voice service response device, wherein the method comprises the following steps: receiving a voice service request, acquiring a voice service function tree corresponding to the voice service, and searching a service processing path for processing the voice service from the voice service function tree, wherein the voice service processing path comprises function nodes required for processing the voice service, and the operation required to be executed for processing the voice service can be obtained according to function trigger operation in the function nodes; and the voice service processing path also comprises a node operation sequence, and the execution sequence of the operation required for processing the voice service can be determined according to the node operation sequence, so that the automatic response to the voice service can be realized according to the service processing path, the operations executed by a user when the voice service is processed are reduced, and the convenience of the user for using the voice service is improved.

Description

Voice service response method and device

The present application claims priority of chinese patent application with application number 201710213583.X entitled "method and apparatus for automatic answering by voice call" filed in 2017, 4/1, the entire contents of which are incorporated herein by reference.

Technical Field

The present application relates to the field of communications technologies, and in particular, to a method and an apparatus for voice service response.

Background

When a user needs a certain service provided by a service provider, a service telephone provided by the service provider is dialed, and service personnel of the service provider provides the service to the user. Since it is costly to provide manual service to users using service personnel and since service personnel are limited, when a large number of users simultaneously request the same service, the speed of providing the service is slow. Therefore, the service provider provides self-service voice service for the user, and can provide the same service for a plurality of users at the same time.

For example, the following steps are carried out: the user needs the service that the communication carrier provides the telephone fee inquiry, the user dials the service telephone (e.g. 10086, etc.) of the communication carrier, after connecting, the communication carrier provides the self-service voice service for the user, the self-service voice service firstly provides the following voice information for the user: chinese service please press 1, For English service press 2, user clicks number 1 according to the above voice information; the self-service voice service then provides the following voice information to the user: the business inquiry is requested to press 1, the mobile phone recharge is requested to press 2, the business handling is requested to press 3, and the user clicks the number 1 according to the voice information; then, the self-service voice service provides the following voice information for the user: the call charge inquiry is requested to be pressed 1, the package and flow inquiry is requested to be pressed 2, the historical bill inquiry is requested to be pressed 3, and the user clicks the number 1 according to the voice information. The self-service voice service provides the service of the call charge inquiry to the user.

According to the above embodiment, when the user uses the self-service voice service, the user needs to perform multiple operations according to the voice information provided by the self-service voice service to process the required service business, and the self-service voice service is complex in operation and poor in convenience.

Disclosure of Invention

In view of this, the present application provides a method and an apparatus for responding to a voice service, which search a service processing path from a voice service function tree corresponding to the voice service, and automatically respond to the voice service by using the service processing path, so as to implement automatic response of the voice service.

A method of voice service answering, the method comprising:

receiving a request for a voice service;

acquiring a voice service function tree of a service provider to which the voice service belongs, wherein the voice service function tree is created according to voice information sent by a server of the service provider;

searching a service processing path from the voice service function tree, wherein the service processing path comprises function nodes and a node operation sequence required by processing the voice service, the function nodes comprise function triggering operations, and the node operation sequence is used for representing an execution sequence of the function triggering operations of the function nodes;

and automatically processing the response of the voice service according to the service processing path.

Optionally, creating a function node in the voice service function tree includes:

responding to a first input operation of a user, and sending a function request corresponding to the first input operation to a server of the service provider;

receiving first voice information sent by the server of the service provider according to the function request;

acquiring a function identifier and a function triggering operation of the first voice information, wherein the function identifier is used for representing a service function of the first voice information, and the function triggering operation is used for representing an operation to be triggered for realizing the service function of the first voice information;

and creating a voice service function tree of a service provider, wherein the voice service function tree comprises a sub-function node corresponding to the first voice information, the sub-function node comprises an identifier of a parent function node, the function identifier and the corresponding relation of the function triggering operation, and the identifier of the parent function node is used for representing the service function which can be realized by the first input operation.

Optionally, the first voice information function triggering operation includes a preset user attribute, and the method further includes:

responding to a second input operation of the user, and acquiring a value of the preset user attribute;

and creating a corresponding relation between the sub-function node and the value of the preset user attribute.

Optionally, the method further includes:

and storing the corresponding relation between the voice service and the service processing path.

Optionally, the obtaining the voice service function tree corresponding to the voice service includes:

performing semantic analysis on the voice service request, and searching whether a service processing path corresponding to the voice service exists;

and when the service processing path corresponding to the voice service does not exist, acquiring a voice service function tree corresponding to the voice service.

Optionally, the automatically processing the response of the voice service according to the service processing path includes:

receiving second voice information sent by the service provider;

acquiring a functional node corresponding to the second voice information from the service processing path, and taking the functional node as a target node;

and processing the second voice information according to the target node.

Optionally, the performing, according to the target node, the operation of processing the second voice information includes:

acquiring a function identifier and a function triggering operation in the target node;

performing semantic analysis on the second voice information, and identifying whether a function identifier of the second voice information is matched with the second voice information;

and when the function identification of the second voice information is matched with the second voice information, executing function triggering operation of the second voice information, and responding to the second voice information.

Optionally, the method further includes:

when the function identification of the second voice information is not matched with the second voice information, the second voice information is played to a user;

and generating a new function node according to the second voice information, and updating the voice service function tree by using the new function node.

An apparatus for voice service answering, the apparatus comprising:

the receiving module is used for receiving a request of voice service;

the first acquisition module is used for acquiring a voice service function tree of a service provider to which the voice service belongs, wherein the voice service function tree is created according to voice information sent by a server of the service provider;

a searching module, configured to search a service processing path from the voice service function tree, where the service processing path includes a function node and a node operation sequence required for processing the voice service, the function node includes a function triggering operation, and the node operation sequence is used to represent an execution sequence of the function triggering operation of the function node;

and the processing module is used for automatically processing the response of the voice service according to the service processing path.

Optionally, the apparatus further comprises:

the sending module is used for responding to a first input operation of a user and sending a function request corresponding to the first input operation to a server of the service provider;

the receiving module is used for receiving first voice information sent by the server of the service provider according to the function request;

a second obtaining module, configured to obtain a function identifier and a function trigger operation of the first voice information, where the function identifier is used to represent a service function of the first voice information, and the function trigger operation is used to represent an operation to be triggered to implement the service function of the first voice information;

the first creating module is used for creating a voice service function tree of a service provider, wherein the voice service function tree comprises a sub-function node corresponding to the first voice information, the sub-function node comprises an identifier of a parent function node, the function identifier and a corresponding relation of function triggering operation, and the identifier of the parent function node is used for representing a service function which can be realized by the first input operation.

Optionally, the first voice information function triggering operation includes a preset user attribute, and the apparatus further includes:

a third obtaining module, configured to obtain a value of the preset user attribute in response to a second input operation of the user;

and the second creating module is used for creating the corresponding relation between the sub-function node and the value of the preset user attribute.

Optionally, the apparatus further comprises:

and the storage module is used for storing the corresponding relation between the voice service and the service processing path.

Optionally, the first obtaining module includes:

the analysis unit is used for carrying out semantic analysis on the voice service request and searching whether a service processing path corresponding to the voice service exists or not;

and the first obtaining unit is used for obtaining the voice service function tree corresponding to the voice service when the service processing path corresponding to the voice service does not exist.

Optionally, the processing module includes:

the receiving unit is used for receiving second voice information sent by the service provider;

a second obtaining unit, configured to obtain a functional node corresponding to the second voice message from the service processing path, and use the functional node as a target node;

and the processing unit is used for processing the second voice information according to the target node.

Optionally, the processing unit includes:

a third obtaining unit, configured to obtain a function identifier and a function trigger operation in the target node;

the analysis unit is used for carrying out semantic analysis on the second voice information and identifying whether a function identifier of the second voice information is matched with the second voice information;

and the execution unit is used for executing the function triggering operation of the second voice information and responding to the second voice information when the function identification of the second voice information is matched with the second voice information.

Optionally, the apparatus further comprises:

the playing unit is used for playing the second voice information to a user when the function identification of the second voice information is not matched with the second voice information;

and the updating unit is used for generating a new function node according to the second voice information and updating the voice service function tree by using the new function node.

An apparatus for voice service answering, the apparatus comprising:

the receiver and the memory are respectively connected with the processor through the communication bus;

the receiver is used for receiving a request of voice service;

the memory to store instructions, the processor to invoke execution of the instructions from the memory, the instructions comprising:

receiving a request for a voice service;

As can be seen from the above, the present application has the following advantages:

receiving a voice service request, acquiring a voice service function tree corresponding to the voice service, and searching a service processing path for processing the voice service from the voice service function tree, wherein on one hand, the voice service processing path comprises a function node required for processing the voice service, and the operation required to be executed for processing the voice service can be obtained according to function trigger operation in the function node; on the other hand, the voice service processing path further comprises a node operation sequence, and the execution sequence of the operation required for processing the voice service can be determined according to the node operation sequence, so that the automatic response to the voice service can be realized according to the service processing path, the operations executed by the user when the voice service is processed are reduced, and the convenience of the user in using the voice service is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments will be briefly introduced below.

Fig. 1 is a flowchart of a method for creating a function node in a voice service function tree according to an embodiment of the present application;

fig. 2 is a schematic diagram of a voice service function tree according to an embodiment of the present application;

fig. 3 is a flowchart of a method for responding to a voice service according to an embodiment of the present application;

fig. 4 is a schematic diagram of a service processing path provided in the embodiment of the present application;

fig. 5 is a flowchart of a method for responding to a voice service according to an embodiment of the present application;

FIG. 6 is a timing diagram illustrating an example scenario of a method for voice response according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a device for responding to a voice service according to an embodiment of the present application;

fig. 8 is a schematic diagram of a hardware structure of a device for responding to a voice service according to an embodiment of the present application.

Detailed Description

In order to provide an implementation scheme of automatic voice service response, embodiments of the present application provide a method and an apparatus for voice service response, and the following describes embodiments of the present application with reference to drawings of the specification.

The user uses a terminal device, such as a mobile phone or the like, to dial a service phone provided by the service provider. After the connection, the server of the service provider provides self-service voice service for the user and sends a section of voice information to the terminal equipment. And the terminal equipment plays the voice information to the user, the user executes input operation according to the voice information, and the terminal equipment returns the input operation to the server of the service provider. And the server of the service provider sends another piece of voice information to the terminal equipment according to the input operation. By analogy, when the terminal device receives a section of voice information returned by the server of the service provider, the user needs to perform an input operation according to the section of voice information until the voice information is the service required by the user. When the service is provided for the user by adopting the mode, the user needs to execute multiple input operations, the operation process is complex, and the user experience is poor.

In order to solve the above problems, the present application provides a voice service response method, which searches a service processing path from a voice service function tree corresponding to a voice service, and implements automatic response of the voice service according to a function node and a node operation sequence in the service processing path, thereby reducing the number of times of executing input operation when a user uses a self-service voice service, simplifying a user operation process, and improving user experience.

The embodiments provided in the present application will be described in detail below with reference to the accompanying drawings.

Before describing the voice service response method provided by the embodiment of the present application, a method for creating a voice service function tree used in the voice service response method is described. The voice service function tree in the embodiment of the present application is created by performing voice recognition on voice information sent by a server of a service provider, and a specific creation method is as follows.

Fig. 1 is a flowchart of a method for creating a function node in a voice service function tree according to an embodiment of the present application, where the method includes:

101: and responding to a first input operation of a user, and sending a function request corresponding to the first input operation to a server of the service provider.

102: and receiving first voice information sent by the server of the service provider according to the function request.

And the user executes a first input operation on a human-computer interaction interface of the terminal equipment. And the terminal equipment receives the first input operation, analyzes a function request corresponding to the first input operation and sends the function request to a server of a service provider. And the server of the service provider searches the first voice information corresponding to the function request from the provided self-service voice service according to the function request, and sends the first voice information to the terminal equipment.

In one case, the sending of the function request corresponding to the first input operation may be implemented by sending a digital code corresponding to the first input operation, where the digital code may be a number or a digital code represented by a Dual Tone Multi Frequency (DTMF) signal. In another case, the function request corresponding to the first input operation may be sent by sending an operation signal corresponding to the first input operation. Of course, in practical application, other modes may be adopted, and the specific setting may be performed according to practical situations, which is not described herein again.

And creating a voice service function tree of the service provider, wherein the voice service function tree of the service provider is not available on the terminal equipment. In an initial state, the function request corresponding to the first input operation of the user is to establish a call connection between the terminal device and the server of the service provider. For example, the first input operation of the user is to dial a service telephone of the service provider on the terminal device. And the terminal equipment responds to the first input operation and sends a function request for establishing call connection to the service of the service provider.

And after establishing call connection with the terminal equipment, the server of the service provider sends first voice information corresponding to the function request to the terminal equipment, wherein the first voice information is the first section of voice information of the self-service voice service.

For example, the following steps are carried out: the user dials 10086 a serving call for chinese mobile on the mobile terminal. The mobile terminal responds to the operation of 10086 of the user, and sends a request for establishing the call connection to the server of the China Mobile. The server mobile terminal of China Mobile establishes a call connection, acquires a first section of voice from self-Service voice Service, welcomes to call China Mobile, and sends a Chinese Service request to terminal equipment according to 1, For English Service, Press 2.

In a non-initial state, the terminal device plays a section of voice information sent by a server of a service provider to a user, the user executes a first input operation according to the voice information, and a function request corresponding to the first input operation is a service function represented by the first input operation in the request of the voice information. The terminal device transmits the function request to a server of a service provider in response to a first input operation by the user. And after receiving the function request, the server of the service provider provides the service function represented by the first input operation. The server of the service provider obtains a section of voice information under the service function from the provided self-service voice service, records the voice information as first voice information, and sends the first voice information to the terminal equipment.

For example: the voice information sent to the terminal equipment by the server of the China Mobile is 'Chinese Service Press 1, For English Service, Press 2', and the first input operation executed on the terminal equipment by the user is 'Press 1' according to the voice information. The terminal equipment sends a function request to the server of China Mobile, namely, the server requesting China Mobile provides Chinese service expressed by '1'. The Chinese mobile server provides Chinese services, and obtains self-service voice services provided by the Chinese mobile, wherein voice information under the Chinese services, namely mobile phone recharging request pressing 1, business query requesting pressing 2 and business handling request pressing 3, is sent to the terminal equipment as first voice information.

103: and acquiring a function identifier and a function triggering operation of the first voice information, wherein the function identifier is used for representing the service function of the first voice information, and the function triggering operation is used for representing the operation to be triggered for realizing the service function of the first voice information.

After receiving the first voice message, the terminal device performs voice recognition on the first voice message by adopting a voice recognition technology, obtains a keyword which is used for evaluating a service function in the first voice message, uses the keyword as a function identifier, obtains a keyword which is used for evaluating an operation to be triggered by realizing the service function in the first voice message, and uses the keyword as a function triggering operation.

Generally, the first voice information is subjected to voice recognition, so that a plurality of function identifiers and a plurality of function triggering operations can be obtained, wherein one function identifier corresponds to one function triggering operation, and a service function represented by one function identifier can only be realized by executing the function triggering operation corresponding to the function identifier. In a specific implementation, the function identifier may be a text or a character string obtained by performing voice recognition on the first voice information.

For example, the following steps are carried out: the first voice message is 'Chinese Service request 1, For English Service, Press 2', and carries out voice recognition to the first voice message. A function identifier 'Chinese' is obtained, and the service function represented by the function identifier is a Chinese service. Another obtained function identifier is "For English Service", and the function identifier represents that the Service function is an English Service. The function trigger operation "press 1" is obtained, which corresponds to the function identifier "Chinese", i.e. the function trigger operation that needs to be executed to implement the Chinese service is "press 1". And obtaining another function triggering operation 'Press 2', wherein the function triggering operation corresponds to a function identifier 'For English Service', namely the function triggering operation For realizing the main execution of the English Service is 'Press 2'.

After receiving the first voice information, if the terminal device is capable of providing a voice recognition technology, the terminal device performs voice recognition on the first voice information by using the voice recognition technology; if the terminal device cannot provide the voice recognition technology, the terminal device can send the first voice information to a voice recognition device. And the voice recognition equipment performs voice recognition on the first voice information and then returns the voice recognition result of the first voice information to the terminal equipment.

104: and creating a voice service function tree of the service provider, wherein the voice service function tree comprises a sub-function node corresponding to the first voice information, the sub-function node comprises a corresponding relation of the identifier of the parent function node, the function identifier and the function trigger operation, and the identifier of the parent function node is used for representing the service function which can be realized by the first input operation.

In the voice service function tree of the service provider, except for the root function node, each function node has a unique father function node corresponding to the function node, and the function node is a child function node of the corresponding father function node.

In the voice service function tree of the service provider, besides leaf sub-function nodes, each function node has at least one sub-function node corresponding to the function node, and the function node is a father function node of the corresponding sub-function node.

The father function node and the son function node are relative concepts, and in the voice service function tree, except for a root node and a leaf node, other function nodes can be used as the father function node and the son function node.

For example, the following steps are carried out: as shown in fig. 2, the root node 201 "china mobile" is the parent function node of the function node 202 "chinese, by 1" and the function node 203 "For English Service, Press 2". That is, the functional node 202 "Chinese, by 1" and the functional node 203 "For English Service, Press 2" are child functional nodes of the root node 201 "China Mobile". The function node 202 "Chinese, by 1" is a child function node of the root node 201 "China Mobile", and is also a parent function node of the function node 204 "Mobile recharge, by 1", the function node 205 "service query, by 2", and the function node 206 "service transaction, by 3".

According to the first input operation, the identifier of the parent function node of the child function node in the voice service function tree of the service provider to be created is obtained, and the identifier of the parent function node may be a function identifier in the parent function node or a self-defined identifier for the parent function node, as long as the parent function node can be identified. The identity of the parent function node corresponds to a service function that can be implemented by the first input operation.

According to the functional service which can be realized by the first input operation, one functional node can be uniquely determined to be used as a father functional node. In the initial state, the functional service which can be realized by the first input operation is the call connection between the terminal equipment and the server of the service provider. At this time, the parent functional node of the child functional node to be established is the root node, and the identifier of the root node is acquired. In a non-initial state, according to the service function which can be realized by the first input operation, a function identifier can be determined, and a function trigger operation corresponding to the function identifier can be determined, so that a created function node can be uniquely determined as a parent function node.

For example, the following steps are carried out: in the initial state, when the first input operation is to dial the service telephone 10086 of the china mobile, the identifier of the parent functional node is the identifier of the root node "china mobile". In a non-initial state, the first input operation is ' press 1 ', the service function realized by the first input operation is ' Chinese service ', a function identifier can be determined to be ' Chinese ', the function trigger operation corresponding to the function identifier is ' press 1 ', a function node ' Chinese can be uniquely determined, according to 1 ', the function node ' Chinese is used as a father function node according to 1 ', and the identifier of the father function node is the identifier of the function node ' Chinese ', according to 1 '.

After the function identifier in the first voice message and the function trigger operation corresponding to the function identifier are obtained, and the identifier of the father function node is obtained, the number of created sub function nodes in the voice service function tree of the service provider is the same as the number of the function identifiers. Each sub-function node comprises the identifier of the parent function node, the function identifier and the corresponding relation of the function triggering operation. The created child function nodes corresponding to the first voice message are all child function nodes of the parent function node. Each sub-function node comprises an identifier of the father function node, a function identifier, a function trigger operation corresponding to the function identifier and a corresponding relation among the three.

It should be noted that, when creating a root node in a voice service function tree, the content of the root node may be set by user. For example, the name of the service provider to which the voice service function tree belongs may be set as a root node, "china mobile"; for another example, the service telephone of the service provider to which the voice service function tree belongs may be set as a root node, root node "10086". Or carrying out voice recognition on the first section of voice information sent by the server of the service provider, and setting the content of the root node according to the voice recognition result. For example, after the first segment of voice information is subjected to voice recognition, a voice part which does not contain a function identifier and a function trigger operation is obtained, the recognition result of the voice part is used as a root node, and the root node 'welcomes to Chinese movement'; for another example, after performing voice recognition on the first segment of voice information, a voice part not including the function identifier and the function trigger operation is obtained, and a keyword is extracted from the recognition result of the voice part as a root node, where the root node is "china move".

Of course, the root node may also be set in other manners according to actual needs, as long as the root node can uniquely correspond to the self-service voice service provided by the service provider.

In one example, the function-triggered operation of the first voice message includes a preset user attribute, and the method further includes:

responding to a second input operation of the user, and acquiring a value of a preset user attribute;

and creating a corresponding relation between the sub-function nodes and the values of the preset user attributes.

If the first voice message includes a keyword representing an operation to be triggered to implement the service function, that is, the function triggering operation includes a keyword representing a preset user attribute, for example: account number, identification card number, etc. And the user executes a second input operation on the user interaction interface of the terminal equipment. And the terminal equipment responds to the second input operation and acquires the value of the preset user attribute in the function triggering operation. And creating a corresponding relation between the sub-function node and the value of the preset user attribute.

When the business processing path includes the sub-function node and the function triggering operation of the sub-function node is executed, the value of the preset user attribute corresponding to the sub-function node is obtained, and the input operation of the value of the preset user attribute is automatically realized.

For example, the following steps are carried out: if the first voice message includes "please input the ID card number, and the ID card number is bound by the # number bond", then obtain "110123199006225678" inputted by the user on the terminal device, create the corresponding relationship between the sub-function node "input ID card number, and the # number bond" of the first voice message and the value "110123199006225678" of the user attribute. When the sub-function node is used, the value "110123199006225678" of the user attribute is acquired, and is automatically input.

Therefore, the user can be prevented from repeatedly inputting the value of the user attribute, the operation of the user is simplified, and the waste of time caused by repeatedly executing the operation due to the input error of the value of the user attribute is reduced.

In the above, a method for creating a functional node corresponding to a piece of voice information after receiving the piece of voice information is described. And creating a voice service function tree of a service provider, wherein a plurality of sections of voice information sent by a server of the service provider needs to be received, and a function node corresponding to each section of voice information is created by adopting the method, so that the voice service function tree of the service provider is finally obtained. By adopting the method, the voice service function trees of a plurality of service providers can be created to realize the automatic response of the voice services of the plurality of service providers.

Therefore, in the voice service response method, the used voice service function tree is created by performing voice recognition on the voice information sent by the server of the service provider based on the voice recognition technology. The voice service function tree is matched with self-service voice service provided by a service provider, and automatic response of the voice service provided by the service provider can be better realized.

After the voice service function tree of the service provider is described in detail, a method for implementing automatic response of the voice service by using the voice service function tree is described in detail below.

Fig. 3 is a flowchart of a method for responding to a voice service according to an embodiment of the present application, where the method includes:

301: a request for voice service is received.

302: and acquiring a voice service function tree of a service provider to which the voice service belongs, wherein the voice service function tree is created according to voice information sent by a server of the service provider.

The user inputs the request of the voice service through a user interaction interface provided by the terminal equipment. The terminal device may provide a voice input function on the user interaction interface, and the user inputs the request of the voice service by using the voice input function. The terminal equipment can also provide a text input function on the user interactive interface, and the user inputs the request of the voice service by using the text input function.

After receiving the voice service request, the terminal equipment firstly performs voice recognition on the voice service request and performs semantic analysis on a text result obtained by voice recognition if the voice service request is voice; if the request of the voice service is a character, the semantic analysis is directly carried out on the request of the voice service, a service provider to which the voice service belongs is analyzed, and a voice service function tree of the service provider is searched.

For example, the following steps are carried out: the request of voice service input by user is "enquire telephone charge balance, China Mobile". The mobile phone carries out semantic analysis on the request of the voice service, and analyzes that the voice service is 'balance inquiry', and the affiliated service provider is 'China Mobile'. The voice service and the service provider to which the voice service belongs can be obtained only by performing semantic analysis on the request of the voice service.

For another example: the voice service request input by the user is 'inquiring the telephone charge balance of the mobile phone', the mobile phone carries out semantic analysis on the voice service request, and the voice service is 'inquiring the balance'. The mobile phone can also automatically analyze that the mobile phone number used by the mobile phone belongs to China Mobile. Therefore, the service provider to which the voice service request of inquiring the telephone charge balance of the mobile phone belongs can be determined to be 'China Mobile'. Performing semantic analysis on the request of the voice service to acquire the voice service; and integrating other information which can be provided by the terminal equipment to determine a service provider to which the voice service belongs.

After the service provider is determined, whether the voice service function tree corresponding to the service provider exists in the mobile terminal is searched, and if the voice service function tree does not exist in the mobile terminal, the voice service function tree of the service provider is created according to the method shown in fig. 1. And if so, acquiring the voice service function tree of the service provider.

303: searching a service processing path from the voice service function tree, wherein the service processing path comprises function nodes and a node operation sequence required by processing the voice service, the function nodes comprise function trigger operations, and the node operation sequence is used for representing the execution sequence of the function trigger operations of the function nodes.

After the voice service function tree of the service provider is obtained, traversing each function node of the voice service function tree from the root node, and searching the function node matched with the voice service semantics. The voice service is semantically matched with the function node, mainly the voice service is matched with the function identification voice of the function node. The function identifier of the function node is used to indicate the service function that the function node can provide. Also, typically, the function node is a leaf node in the voice service function tree.

The voice service function tree of the service provider can represent the processing logic of the self-service voice service provided by the service provider. After the functional node matched with the voice service semantics is found, the only service processing path from the root node to the functional node can be found from the service functional tree. The service processing path comprises functional nodes required for processing the voice service and also comprises a node operation sequence. The node operation sequence is used for representing the execution sequence of the function triggering operation of each functional node when the service processing path processes the voice service. It is understood that, in general, the traffic processing path does not include the root node in the voice traffic function tree. In the service processing path, each functional node corresponds to a section of voice information sent by a service provider when the service provider provides the voice service.

And when the service processing path is found, the terminal equipment automatically establishes voice connection with the server of the service provider.

Taking the voice service function tree shown in fig. 2 as an example, the following are illustrated: the voice service request input by the user is 'inquiring the balance of the phone charge of the mobile phone', the voice service obtained by analysis is 'inquiring the balance', and the service provider is 'China Mobile'. Acquiring a voice service function tree of China Mobile, and searching a function node matched with the 'balance inquiry' semantic from the voice service function tree, wherein the function node is 'balance inquiry, and is according to 1'. From the voice service function tree, a service processing path from a root node ' welcome to China Mobile ' to a function node ' balance inquiry ' is searched according to 1 '. As shown in fig. 4, the service processing path includes functional nodes in sequence as follows: the functional node 'Chinese, according to 1', the functional node 'service inquiry, according to 2', the functional node 'balance inquiry, according to 1'. The service processing path shown in fig. 4 includes both the functional nodes required for processing the voice service of "query balance", and the function triggering operations of the respective functional nodes, and also includes the execution sequence "by 1, by 2, by 1" of the function triggering operations of the respective functional nodes.

304: and automatically processing the response of the voice service according to the service processing path.

And after the service processing path is obtained, automatically processing the response of the voice service according to the execution sequence of the function triggering operation of each functional node in the service processing path.

And the terminal equipment establishes voice connection with the server of the service provider, receives second voice information sent by the server of the service provider, acquires a function node corresponding to the second voice information from the service processing path as a target node, and processes the second voice information by using the target node.

In one example, the traffic processing path includes only function-triggered operations of the functional node. When the response of the voice service is automatically processed according to the service processing path, the terminal equipment acquires the function triggering operation of a function node from the service processing path according to the sequence every time the terminal equipment receives second voice information sent by a server of a service provider, executes the function triggering operation of the function node, and automatically responds to the received second voice information.

For example, the following steps are carried out: the terminal device finds out the service processing path from the voice service function tree as follows: "press 1, press 2, press 1". The terminal equipment establishes voice connection with the China Mobile server, receives a first section of voice information sent by the China Mobile server, acquires a function triggering operation 'press 1' of a first function node from a business processing path, and automatically executes the operation according to 1; the server of the China Mobile sends a second section of voice information to the terminal equipment according to the operation according to 1, the terminal equipment receives the second section of voice information, obtains the function triggering operation 'according to 2' of the second functional node in sequence from the service processing path, and automatically executes the operation according to 2; the server of the China Mobile sends a third section of voice information to the terminal equipment according to the operation according to 2, the terminal equipment receives the third section of voice information, obtains the function triggering operation 'according to 1' of a third functional node in sequence from the service processing path, and automatically executes the operation according to 1; and the China mobile server acquires the balance of the mobile phone according to the operation 1 and returns the balance to the mobile terminal. The balance of the mobile phone can be sent to the terminal equipment in a voice mode and played to the user by the terminal equipment; or the data can be sent to the terminal equipment in a short message form and displayed to the user by the terminal equipment.

In another example, the traffic processing path includes not only the function triggering operation of the function node, but also the function identification of the function node. When the response of the voice service is automatically processed according to the service processing path, the terminal equipment acquires a functional node from the service processing path according to the sequence after receiving second voice information sent by a server of a service provider, compares the received second voice information with the semantics of the functional identifier of the functional node to determine whether the functional node is the functional node of the service function responding to the second voice information. If the semanteme is matched, the function node is the function node of the service function responding to the second voice information, the function triggering operation of the function node is executed, and the second voice information is automatically responded.

If the semantics are not matched, the functional node is not the functional node of the service function for responding the second voice message, the terminal equipment plays the second voice message to the user, and the user executes the operation for responding the second voice message. Meanwhile, by using the method shown in fig. 1, a new function node is generated according to the second voice message, and the voice service function tree of the service provider is updated by using the new function node.

In this example, the received second voice information is subjected to voice recognition, and the voice recognition result of the second voice information is compared to determine whether the semantic meaning of the function identifier of the function node is matched with the semantic meaning of the function identifier of the function node, so that the accuracy of automatic response of the voice service can be improved, and an automatic response error can be avoided when the processing logic of the self-service voice service provided by a service provider is changed and the terminal device is only triggered to operate according to the functions of the function nodes in the service processing path and automatically responds according to the sequence of the function triggering operations. And, when the processing logic of the self-service voice service provided by the service provider is changed, the voice service function tree of the service provider can be automatically updated.

For example, the following steps are carried out: the terminal device finds out the service processing path from the voice service function tree as follows: chinese, press 1 "," service inquiry, press 2 "," balance inquiry, press 1 ". The method comprises the steps that a voice connection is established between terminal equipment and a China Mobile server, the terminal equipment receives a first section of voice information sent by the China Mobile server, a first functional node Chinese is obtained from a business processing path and is pressed according to 1, whether the semantics of the first section of voice information and the Chinese of a functional identifier of the first functional node are matched or not is compared, if the semantics are matched, a function trigger operation 'pressing 1' of the first functional node is obtained, and the operation according to 1 is automatically executed. The server of the China Mobile sends a second section of voice information to the terminal equipment according to the operation according to 1, the terminal equipment receives the second section of voice information, acquires a second functional node ' service query ' from the service processing path according to 2 ', firstly compares whether the second section of voice information is matched with the ' service query ' semantic of the functional identifier of the second functional node, if so, acquires the function triggering operation ' press 2 ' of the second functional node, and automatically executes the operation according to 2. The China Mobile server sends a third section of voice information to the terminal equipment according to the operation according to the 2, the terminal equipment receives the third section of voice information, acquires a third functional node balance inquiry from the service processing path, according to the 1, compares whether the third section of voice information is matched with the semantic meaning of the function identifier balance inquiry of the third functional node, if so, acquires the function triggering operation of the third functional node 'according to the 1', automatically executes the operation according to the 1, and the China Mobile server provides a balance inquiry result for the terminal equipment. If the third section of voice information is that the local machine service inquires and asks 1 and other mobile phones inquire and ask 2, the third section of voice information is not matched with the function identifier 'balance inquiry' of the third function node, the third section of voice information is played to the user, and the user executes response operation. And, according to the method shown in fig. 1, a new functional node is generated according to the third segment of voice information, and the voice service functional tree of the chinese mobile is updated by using the new functional node.

Based on the method for creating the voice service function tree of the service provider shown in fig. 1, a method for automatically responding to the voice service shown in fig. 3 is provided, a service processing path for processing the voice service is obtained from the voice service function tree created in advance, and according to the function triggering operation of the function node in the service processing path and the execution sequence of the function triggering operation, the automatic response to the voice service is realized, the operations executed by the user when processing the voice service are reduced, and the convenience of using the voice service by the user is improved.

In another example, the method for responding to the voice service further includes storing a corresponding relationship between the voice service and the service processing path, and before acquiring the voice service function tree of the service provider, first searching whether the service processing path corresponding to the voice service exists, and if so, directly acquiring the service processing path corresponding to the voice service without acquiring the voice service function tree of the service provider. As shown in fig. 5, the method comprises:

501: a request for voice service is received.

502: inquiring whether a service processing path corresponding to the voice service exists, if so, executing 506; if not, 503 is performed.

503: and acquiring a voice service function tree of a service provider to which the voice service belongs, wherein the voice service function tree is created according to voice information sent by a server of the service provider.

504: searching a service processing path from the voice service function tree, wherein the service processing path comprises function nodes and a node operation sequence required by processing the voice service, the function nodes comprise function trigger operations, and the node operation sequence is used for representing the execution sequence of the function trigger operations of the function nodes.

505: and storing the corresponding relation between the voice service and the service processing path.

506: and automatically processing the response of the voice service according to the service processing path.

Step 501, step 503, step 504, and step 506 are similar to the method shown in fig. 3, and refer to the description of the method shown in fig. 3, which is not described herein again.

After receiving the request of the voice service, the voice service function tree of the service provider to which the voice service belongs is not immediately acquired, but whether a service processing path corresponding to the voice service exists is firstly searched. If the service processing path exists, the voice service function tree of the service provider does not need to be acquired, and the response of the voice service can be automatically processed by directly utilizing the service processing path corresponding to the voice service. If the service processing path corresponding to the voice service is not found, the voice service function tree of the service provider is obtained, and the service processing path for processing the voice service is found from the voice service function tree.

It should be noted that, in general, when searching for a service processing path corresponding to a voice service, a request of the voice service input by a user needs to be semantically matched with a stored voice service, so as to find the service processing path corresponding to the voice service. Or, the user is required to input the voice service according to the standard expression of the voice service, so that the accuracy of searching the service processing path corresponding to the voice service is improved.

It can be understood that, in the method shown in fig. 5, compared with the method shown in fig. 3, if there is a service processing path corresponding to the voice service, that is, the service processing path is not searched from the voice service function tree of the service provider, the speed of automatically responding to the voice service is further increased.

To better illustrate the implementation scheme provided by the present application, a scenario example of the voice response method provided by the present application is given below, as shown in fig. 6, where a user requests to query a balance:

601: the user inputs a request of voice service of 'inquiring the balance of the mobile phone' on the mobile phone.

602: the mobile phone inquires whether a service processing path corresponding to the voice service of 'inquiring the balance of the mobile phone' exists, if so, 604 is executed; if not, 603 is performed.

603: the mobile phone obtains the voice service function tree of the Chinese mobile according to the voice service of 'inquiring balance' and the communication operator to which the mobile phone number belongs, and obtains the service processing path from the voice service function tree of the Chinese mobile.

604: the mobile phone establishes voice connection with the server of the China Mobile.

605: the China Mobile Server sends the first voice message "welcome to call China Mobile, Chinese Service Press 1, For English Service, Press 2" to the handset.

606: and the mobile phone identifies the function identifier 'Chinese' of the first function node in the service processing path, and executes the function triggering operation 'press 1' of the first function node if the function identifier 'Chinese' is matched with the first section of voice information.

607: the server of the China Mobile sends a second section of voice information to the mobile phone according to the operation of the mobile phone 1, wherein the mobile phone recharge request is according to 1, the service query request is according to 2, and the service transaction request is according to 3.

608: and in the service processing path identified by the mobile phone, the function identifier 'service inquiry' of the second function node is matched with the second section of voice information, and then the function triggering operation 'press 2' of the second function node is executed.

609: the server of the China Mobile sends a third section of voice information 'balance inquiry request is pressed as 1 and package balance inquiry request is pressed as 2' to the mobile phone according to the operation of pressing 2.

610: and in the service processing path identified by the mobile phone, the function identifier 'balance inquiry' of the third function node is matched with the third section of voice information, and then the function triggering operation 'press 1' of the third function node is executed.

611: the service of China Mobile sends balance information to the mobile phone according to the operation of 1.

612: the mobile phone feeds back the balance information to the user.

The example shown in fig. 6 is a scene example of the method for responding to the voice service provided by the present application, and the specific implementation manner is similar to the method shown in fig. 3 and 5, and reference is made to the description in the method shown in fig. 3 and 5, which is not described again here.

Fig. 7 is a schematic structural diagram of a device for responding to a voice service according to an embodiment of the present application, where the device includes:

a receiving module 701, configured to receive a request of a voice service.

A first obtaining module 702, configured to obtain a voice service function tree of a service provider to which the voice service belongs, where the voice service function tree is created according to voice information sent by a server of the service provider.

A searching module 703, configured to search a service processing path from the voice service function tree, where the service processing path includes a function node and a node operation sequence required for processing the voice service, the function node includes a function triggering operation, and the node operation sequence is used to represent an execution sequence of the function triggering operation of the function node.

A processing module 704, configured to automatically process the response of the voice service according to the service processing path.

Optionally, the apparatus further comprises:

Optionally, the first obtaining module 702 includes:

Optionally, the processing module 704 includes:

Optionally, the processing unit includes:

Optionally, the apparatus further comprises:

The apparatus shown in fig. 7 is a device corresponding to the method shown in fig. 1 to 6, and the specific implementation method is similar, and reference is made to the description of the method shown in fig. 1 to 6, which is not repeated here.

Fig. 8 is a schematic diagram of a hardware structure of a voice service response device according to an embodiment of the present application, where the hardware structure includes:

the receiver 801, the processor 802, the memory 803, the communication bus 804, the receiver 801 and the memory 803 are respectively connected with the processor 802 through the communication bus 804.

The receiver 801 is configured to receive a request for a voice service.

The memory 803 is used for storing instructions, and the processor 802 is used for calling the instructions from the memory 803 to execute, wherein the instructions comprise:

receiving a request for a voice service;

The apparatus shown in fig. 8 is a device corresponding to the method shown in fig. 1 to 6, and the specific implementation method is similar, and reference is made to the description of the method shown in fig. 1 to 6, which is not repeated here.

Claims

1. A method of voice service answering, the method comprising:

receiving a request for a voice service;

searching a function node matched with the voice service from a root node in the voice service function tree through traversing function nodes, and determining a service processing path in the voice service function tree according to the function node matched with the voice service, wherein the service processing path comprises function nodes and a node operation sequence required by processing the voice service, the function nodes comprise function triggering operations, and the node operation sequence is used for representing an execution sequence of the function triggering operations of the function nodes;

and automatically processing the response of the voice service according to the execution sequence of the function triggering operation of the function node matched with the voice service in the service processing path.

2. The method of claim 1, wherein creating a function node in the voice service function tree comprises:

3. The method of claim 2, wherein the first voice information function trigger operation comprises a preset user attribute, and wherein the method further comprises:

4. The method of claim 1, further comprising:

5. The method according to claim 4, wherein the obtaining the voice service function tree corresponding to the voice service comprises:

6. The method according to any of claims 1-5, wherein said automatically processing the response of the voice traffic according to the traffic processing path comprises:

receiving second voice information sent by the service provider;

and processing the second voice information according to the target node.

7. The method of claim 6, wherein the performing the operation of processing the second voice information according to the target node comprises:

8. The method of claim 7, further comprising:

9. An apparatus for responding to a voice service, the apparatus comprising:

the receiving module is used for receiving a request of voice service;

a searching module, configured to search, from a root node in the voice service function tree, a function node matched with the voice service through traversing function nodes, and determine a service processing path in the voice service function tree according to the function node matched with the voice service, where the service processing path includes a function node and a node operation sequence required for processing the voice service, the function node includes a function triggering operation, and the node operation sequence is used to represent an execution sequence of the function triggering operation of the function node;

and the processing module is used for automatically processing the response of the voice service according to the execution sequence of the function triggering operation of the function node matched with the voice service in the service processing path.

10. The apparatus of claim 9, further comprising:

11. The apparatus of claim 10, wherein the first voice message function trigger operation comprises a preset user attribute, and wherein the apparatus further comprises:

12. The apparatus of claim 9, further comprising:

13. The apparatus of claim 12, wherein the first obtaining module comprises:

14. The apparatus according to any one of claims 9-13, wherein the processing module comprises:

15. The apparatus of claim 14, wherein the processing unit comprises:

16. The apparatus of claim 15, further comprising:

17. An apparatus for responding to a voice service, the apparatus comprising:

the receiver is used for receiving a request of voice service;

receiving a request for a voice service;

18. An apparatus for voice service answering, comprising a memory, one or more processors, and one or more programs; wherein the one or more programs are stored in the memory; wherein the one or more processors, when executing the one or more programs, cause the apparatus for responding to voice traffic to implement the method of any of claims 1-8.

19. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the method according to any one of claims 1-8.