CN112199498A - Man-machine conversation method, device, medium and electronic equipment for endowment service - Google Patents
Man-machine conversation method, device, medium and electronic equipment for endowment service Download PDFInfo
- Publication number
- CN112199498A CN112199498A CN202011034916.0A CN202011034916A CN112199498A CN 112199498 A CN112199498 A CN 112199498A CN 202011034916 A CN202011034916 A CN 202011034916A CN 112199498 A CN112199498 A CN 112199498A
- Authority
- CN
- China
- Prior art keywords
- endowment
- natural language
- slot
- language text
- endowment service
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000012545 processing Methods 0.000 claims abstract description 19
- 230000009471 action Effects 0.000 claims description 31
- 238000012706 support-vector machine Methods 0.000 claims description 11
- 239000013598 vector Substances 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 5
- 230000000306 recurrent effect Effects 0.000 claims description 5
- 238000009432 framing Methods 0.000 claims description 4
- 230000032683 aging Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 239000002609 medium Substances 0.000 description 14
- 230000008569 process Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 230000004044 response Effects 0.000 description 7
- 238000012790 confirmation Methods 0.000 description 5
- 230000003993 interaction Effects 0.000 description 4
- 230000006872 improvement Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005352 clarification Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000012120 mounting media Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
- G06F40/35—Discourse or dialogue representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Machine Translation (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Human Resources & Organizations (AREA)
Abstract
The embodiment of the application discloses a man-machine conversation method, a man-machine conversation device, a man-machine conversation medium and electronic equipment for endowment services. The method is performed by an electronic device, the method comprising: receiving voice content of a user, and converting the voice content into a natural language text; determining a target service type according to the natural language text; filling slot position information by adopting a slot position type corresponding to the target service type; and generating an endowment service order according to the filled slot position information, and executing the endowment service order. By means of the scheme, the old-age care service requirements can be rapidly identified and processed through a man-machine conversation method, so that the purposes of improving the processing efficiency of the old-age care service requirements and reducing the labor cost of the old-age care service are achieved.
Description
Technical Field
The embodiment of the application relates to the technical field of human-computer interaction, in particular to a human-computer conversation method, a human-computer conversation device, a human-computer conversation medium and electronic equipment for endowment services.
Background
At present, as the global population enters the aging stage, the endowment service becomes a popular topic of the whole society.
At present, most of the demands for old-age services are difficult to be met by the old. The endowment service requirement mainly depends on manual mode (call center staff, community staff, endowment service mechanism service staff and the like) to replace the operation of old people, the human cost is high, the efficiency is low, and in addition, the problems that the endowment service requirement response speed is low and the order forming process is complex can be caused.
Disclosure of Invention
The embodiment of the application provides a man-machine conversation method, a man-machine conversation device, a medium and electronic equipment for endowment services, and the man-machine conversation method, the man-machine conversation device, the medium and the electronic equipment can rapidly identify and process the endowment service requirements through the man-machine conversation means, so that the purposes of improving the efficiency of processing the endowment service requirements and reducing the labor cost of the endowment services are achieved.
In a first aspect, an embodiment of the present application provides a human-computer conversation method for an aging service, where the method includes:
receiving voice content of a user, and converting the voice content into a natural language text;
determining a target service type according to the natural language text;
filling slot position information by adopting a slot position type corresponding to the target service type;
and generating an endowment service order according to the filled slot position information, and executing the endowment service order.
Further, after determining the target service type, the method further includes:
determining a user intention from the target service type according to the natural language text;
correspondingly, generating an endowment service order according to the filled slot position information, and executing the endowment service order, wherein the endowment service order comprises the following steps:
and generating an endowment service order according to the filled slot position information and the user intention, and executing the endowment service order.
Further, receiving the voice content of the user, and converting the voice content into a natural language text, including:
performing effective information extraction pretreatment on the voice content to obtain effective voice content with silence at two ends removed;
performing framing processing on the effective voice content to obtain an effective voice frame;
and processing the effective voice frame to obtain a natural language text.
Further, processing the valid speech frame to obtain a natural language text, including:
performing characteristic extraction on the effective speech frame through a linear prediction cepstrum coefficient;
and decoding the extracted features by adopting an acoustic model and/or a language model to obtain a natural language text.
Further, the acoustic model comprises a hidden markov model; the language model includes an n-gram model.
Further, the target service type includes: at least one of an endowment service order, an endowment service evaluation, a community endowment campaign, and an endowment service recommendation.
Further, determining a target service type according to the natural language text, including:
extracting corpus feature vectors of the natural language text by adopting a TF-IDF algorithm;
and inputting the corpus feature vector into a support vector machine, and determining a target service type according to an output result of the support vector machine.
Further, filling the slot information with the slot type corresponding to the target service type includes:
and filling slot position information by adopting a recurrent neural network model according to the natural language text.
Further, after filling the slot information with the slot type corresponding to the target service type and after determining the user intention from the target service type according to the natural language text, the method further includes:
and determining to execute an action according to the slot position information and the user intention.
Further, the performing act includes: executing tasks, clarifying slot positions, confirming slot values and failing matching.
Further, if the execution action is to execute a service, executing:
and generating an endowment service order according to the filled slot position information and the user intention, and executing the endowment service order.
Further, if the execution action is to clarify the slot position, a question statement is sent out according to the slot value information of the slot position;
and if the execution action is to confirm the slot value, sending the information of the slot value to be confirmed to the user for the user to confirm.
In a second aspect, an embodiment of the present application provides a human-machine conversation device for an endowment service, the device including:
the natural language text conversion module is used for receiving the voice content of a user and converting the voice content into a natural language text;
the target service type determining module is used for determining a target service type according to the natural language text;
the slot position information filling module is used for filling the slot position information by adopting a slot position type corresponding to the target service type;
and the order generation module is used for generating an endowment service order according to the filled slot position information and executing the endowment service order.
In a third aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a man-machine conversation method of an endowment service according to an embodiment of the present application.
In a fourth aspect, an embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable by the processor, where the processor implements a man-machine conversation method of an endowment service according to an embodiment of the present application when executing the computer program.
According to the technical scheme provided by the embodiment of the application, the voice content of a user is received, and the voice content is converted into a natural language text; determining a target service type according to the natural language text; filling slot position information by adopting a slot position type corresponding to the target service type; and generating an endowment service order according to the filled slot position information, and executing the endowment service order. The technical scheme that this application provided can be through man-machine conversation's means, and is quick discernment and processing to endowment service demand to reach the improvement and to endowment service demand treatment effeciency, and reduce the purpose of the human cost of endowment service.
Drawings
FIG. 1 is a flowchart of a man-machine conversation method of an endowment service provided by an embodiment of the application;
FIG. 2 is a diagram illustrating a man-machine conversation method of another endowment service provided by an embodiment of the present application;
FIG. 3 is a schematic diagram of a human-machine conversation system module of an endowment service provided by an embodiment of the application;
FIG. 4 is a schematic structural diagram of a human-machine conversation device of an endowment service provided by an embodiment of the application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be further noted that, for the convenience of description, only some of the structures related to the present application are shown in the drawings, not all of the structures.
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the steps as a sequential process, many of the steps can be performed in parallel, concurrently or simultaneously. In addition, the order of the steps may be rearranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
Fig. 1 is a flowchart of a man-machine conversation method of an endowment service provided in an embodiment of the present application, where the present embodiment is applicable to an endowment service situation, and the method can be executed by a man-machine conversation device of an endowment service provided in an embodiment of the present application, and the device can be implemented by software and/or hardware, and can be integrated in an electronic device such as an intelligent terminal.
As shown in fig. 1, the man-machine conversation method of the endowment service includes:
s110, receiving the voice content of the user, and converting the voice content into a natural language text.
The user can be the old, or the child of the old, and the voice content of the user is received, the voice content can be received after the old calls the fixed number through the mobile terminal, or the voice content can be obtained through other modes, for example, the voice information sent by the old is obtained through the public number of the WeChat, or the voice content is obtained through the front-end voice obtaining device installed in the home of the old.
And converting the voice content into a natural language text, wherein the natural language text can be character content corresponding to the voice, and also can be a character string obtained by coding the language by adopting some means.
In this embodiment, optionally, receiving a voice content of a user, and converting the voice content into a natural language text, includes:
performing effective information extraction pretreatment on the voice content to obtain effective voice content with silence at two ends removed;
performing framing processing on the effective voice content to obtain an effective voice frame;
and processing the effective voice frame to obtain a natural language text.
The valid voice content may be obtained by removing the mute content at the front end and the back end of the language. After the silence is removed, the voice can be framed according to the time sequence to obtain the voice information of each frame. Wherein the division may be performed in the number of 25 frames or 30 frames per second. After obtaining the effective speech frame, the effective speech frame can be further identified to obtain a natural language text. By means of the scheme, the text content of the user voice can be obtained more accurately.
In this scheme, optionally, the valid speech frame is processed to obtain a natural language text, including:
performing characteristic extraction on the effective speech frame through a linear prediction cepstrum coefficient;
and decoding the extracted features by adopting an acoustic model and/or a language model to obtain a natural language text.
Specifically, mute at two ends is cut off, and sound framing is completed. Then, feature extraction is carried out through Linear Prediction Cepstrum Coefficients (LPCC) to obtain a multi-bit vector containing sound information:
wherein, { a1,a2,…,apLPC coefficient; h (n) is a linear prediction cepstrum coefficient.
Here, a Linear Prediction Cepstrum Coefficient (LPCC) is a representation of a Linear Prediction Coefficient (LPC) in the cepstral domain. The feature is based on the value of the speech signal as an autoregressive signal, and cepstral coefficients are obtained by linear predictive analysis.
Next, the features are decoded according to an acoustic model, a language model, and the like.
In this scheme, optionally, the acoustic model includes a hidden markov model; the language model includes an n-gram model.
The acoustic model adopts a Hidden Markov Model (HMM):
P(Y)＝∑XP(Y|X)P(X)；
Y＝y(0),y(1),…,y(l-1)；
X＝x(0),x(1),…,x(l-1)；
in the formula, the observed result is Y, the hiding condition is X, and the length is l;
the language model adopts an N-gram model, and supposing that the output of a certain word is only related to the probability of the occurrence of the previous N-1 words, the model formula is as follows:
P(w)＝P(w1)P(w2|w1)P(w3|w1,w2)…P(wn|w1,w2,…,wn-1)；
after decoding according to the two models and the dictionary, natural language text is generated and used as input of the field recognition step.
According to the scheme, the model which is skillfully used is adopted, the effect of improving the accuracy of natural language text recognition can be achieved, and the requirement for positioning the old people service can be met more accurately.
And S120, determining the target service type according to the natural language text.
Wherein, aiming at different service requirements of the old people, the target service type can be determined. It can be understood that the service types may include a plurality of types, and it may be determined which service type is required for the current voice content of the elderly through recognition of the natural language text, where the determined service type is the target service type.
In this embodiment, optionally, the target service type includes: at least one of an endowment service order, an endowment service evaluation, a community endowment campaign, and an endowment service recommendation. The service type of the endowment service order and the service type of the endowment service evaluation are related to the endowment service in the actual use process, and can be service functions of order placing, order query, order area evaluation on completed orders and the like. Community endowment activities and endowment service recommendations are generally less, but may be retained as one of the classifications of voice information for the elderly in order to meet the actual needs of the elderly.
In this embodiment, optionally, determining the target service type according to the natural language text includes:
extracting corpus feature vectors of the natural language text by adopting a TF-IDF algorithm;
and inputting the corpus feature vector into a support vector machine, and determining a target service type according to an output result of the support vector machine.
Specifically, the process of determining the target service type is as follows:
1. and (3) extracting the characteristics of the user input corpus by using a TF-IDF algorithm:
tfiidfi,j＝tfi,j*idfi,j；
wherein n isi,jIs the word tiIn document djThe number of occurrences, | D | is the total number of documents in the corpus, | { j: t |, in the corpusi∈djDenotes a file djThe sum of the occurrence times of all words in (b). tfiidf (tm) fi,jRepresenting the weight of the word.
2. And inputting the feature vector of the previous step into a Support Vector Machine (SVM) model according to a preset endowment service field, and outputting field classification. Objective function of SVM:
wherein m is the number of samples, w and gamma are respectively the hyperparameters of the SVM model, and the optimal solution of the optimization function is solved, thereby realizing classification
By adopting the TF-IDF algorithm to extract the characteristic vector and adopting the support vector machine to finish the classification of the service types, the aim of accurately determining the types of the users according to the requirements of the users without manual intervention can be fulfilled.
And S130, filling the slot position information by adopting the slot position type corresponding to the target service type.
The slot type may be corresponding to the target service type, that is, if there are four service types, there may be four slot types. The slot information required by each slot type may be different, such as slot definition of an endowment service order, which may include service items, service quantity, reservation time, payment method, and user number.
In this technical solution, optionally, the filling the slot information with the slot type corresponding to the target service type includes:
and filling slot position information by adopting a recurrent neural network model according to the natural language text.
Slot filling is done using a Recurrent Neural Network (RNN) model. The RNN model is defined as follows:
ht＝σ(Wihxt+bih+Whhht-1+bhh)；
xtis the input of the current time, ht-1Is the stealth output at the last moment.
The scheme can adopt the recurrent neural network to fill the slot position quickly and accurately
And S140, generating an endowment service order according to the filled slot position information, and executing the endowment service order.
If the slot information is identified to be filled completely or the filling can not be continued, an endowment service order can be generated according to the slot information. And executes the endowment service order.
On the basis of the above technical solution, optionally, after determining the target service type, the method further includes:
determining a user intention from the target service type according to the natural language text;
correspondingly, generating an endowment service order according to the filled slot position information, and executing the endowment service order, wherein the endowment service order comprises the following steps:
and generating an endowment service order according to the filled slot position information and the user intention, and executing the endowment service order.
Here, the user input intention can be recognized and recorded, and different intentions can be matched with different slot positions; for example, for the field of endowment service orders, the user's intentions include: generating a new order, canceling an existing order and checking order information. The intention detection module has the same processing flow as the field identification module, but has different input and output. It can be understood that the actual intention of the user can be more accurately grasped by adding the intention identification process, and the experience of the user is improved.
In this scheme, optionally, after filling the slot information with the slot type corresponding to the target service type and after determining the user intention from the target service type according to the natural language text, the method further includes:
and determining to execute an action according to the slot position information and the user intention.
The determined execution action may include continuing to confirm, or having confirmed completion, or failing to confirm, and the like, and the execution action may perform a corresponding operation, such as continuing to return information to the user for further confirmation, or determining that information confirmation is completed, and may generate an order according to the current information, or directly determining that information confirmation fails, and an order cannot be placed, and the like.
On the basis of the foregoing technical solution, optionally, the executing act includes: executing tasks, clarifying slot positions, confirming slot values and failing matching.
Specifically, if the execution action is an execution service, the following steps are executed:
and generating an endowment service order according to the filled slot position information and the user intention, and executing the endowment service order.
If the execution action is to clarify the slot position, sending a question sentence according to the slot value information of the slot position;
and if the execution action is to confirm the slot value, sending the information of the slot value to be confirmed to the user for the user to confirm.
Here, the intelligent endowment service processing may be performed based on the recorded slot filling and intention identification information. Such as generating an endowment service order for the user, recording an endowment service rating, and so forth.
Otherwise, according to the template side, converting the slot filling information, the to-be-executed dialogue action and other structural information into natural language description, generating voice and feeding back to the user. In this step, a template-based method is used for feedback. Specifically, the following operation processes may be adopted:
and step 1, feeding back the combination according to the conversation state, the action to be executed and the slot value. If the action to be executed is used as a clarification slot position or a confirmation slot value, executing voice recognition and continuously determining response information; if the action to be executed is 'task completion', executing a corresponding task; if the action to be performed is taken as 'matching failure', executing step 4;
step 2, matching the template and filling the content according to the slot position, the action and the state to generate response information; after the voice content is fed back to the user, the voice content sent by the user is continuously received and recognized;
step 3, according to the determined execution action, executing a success or failure result, matching the template and filling the content, and ending the conversation process after generating response information;
and 4, returning an unidentifiable response.
According to the technical scheme, a user accesses to the endowment service dialogue system and inputs voice content; firstly, the system carries out voice recognition and converts the voice into a text; then, completing field identification, slot filling and intention identification to confirm the user endowment service requirement content; defining subsequent system actions through dialog state management; and generating a voice reply based on the action and the template, finally automatically completing corresponding endowment service business operation, and ending the conversation.
According to the technical scheme provided by the embodiment of the application, the voice content of a user is received, and the voice content is converted into a natural language text; determining a target service type according to the natural language text; filling slot position information by adopting a slot position type corresponding to the target service type; and generating an endowment service order according to the filled slot position information, and executing the endowment service order. The technical scheme that this application provided can be through man-machine conversation's means, and is quick discernment and processing to endowment service demand to reach the improvement and to endowment service demand treatment effeciency, and reduce the purpose of the human cost of endowment service.
Fig. 2 is a schematic diagram of a man-machine conversation method of another endowment service provided by an embodiment of the present application, as shown in fig. 2, the method includes:
a user actively initiates a conversation to the system and sends voice content;
after receiving the voice content, the system converts the voice content into a natural language text;
identifying the specific field to which the user input voice belongs, and guiding the user to confirm to enter the functional module of the corresponding field;
recognizing and recording user input intentions, wherein different intentions can be matched with different slot positions;
receiving slot filling information and an intention identification result, integrating conversation information and a historical state, updating a current conversation state, and producing a conversation action to be executed;
if the action to be executed is 'task execution', intelligent endowment service processing is carried out according to the recorded slot filling and intention identification information;
according to the template side, converting slot filling information, to-be-executed dialogue action and other structured information into natural language description, generating voice and feeding the voice back to a user; specifically, the method comprises the following steps: i. and feeding back the combination according to the conversation state, the action to be executed and the slot value. If the action to be executed is used as a clarification slot position and a confirmation slot value, if the action to be executed is used as a completion task, the successful or failed execution result is matched with the template and filled with the content, and after response information is generated, the conversation process is ended; if the pending action is taken as "match failure," a no-recognition response is returned.
The scheme is applied to the processing flow of the endowment service field through a man-machine conversation system: a user accesses the endowment service dialogue system and inputs voice content; firstly, the system carries out voice recognition and converts the voice into a text; then, completing field identification, slot filling and intention identification to confirm the user endowment service requirement content; defining subsequent system actions through dialog state management; and generating a voice reply based on the action and the template, finally automatically completing corresponding endowment service business operation, and ending the conversation.
Fig. 3 is a schematic diagram of a human-machine conversation system module of an endowment service provided in an embodiment of the present application, and as shown in fig. 3, the human-machine conversation system of the endowment service includes: the system comprises a user access module, a voice recognition module, a field recognition module, an intention detection module, a slot filling module, a conversation management module, an intelligent endowment service processing module and a reply processing module.
The system can carry out semantic analysis and understanding on the voice of the customer, identify the endowment service requirement of the customer and intelligently process the endowment service business
Fig. 4 is a schematic structural diagram of a man-machine interaction device of an endowment service provided by an embodiment of the present application. As shown in fig. 4, the man-machine conversation device of the endowment service includes:
a natural language text conversion module 410, configured to receive a voice content of a user, and convert the voice content into a natural language text;
a target service type determining module 420, configured to determine a target service type according to the natural language text;
a slot position information filling module 430, configured to fill slot position information with a slot position type corresponding to the target service type;
the order generating module 440 is configured to generate an endowment service order according to the filled slot position information, and execute the endowment service order.
The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method.
Embodiments of the present application also provide a storage medium containing computer-executable instructions that, when executed by a computer processor, perform a human-machine conversation method for an endowment service, the method comprising:
receiving voice content of a user, and converting the voice content into a natural language text;
determining a target service type according to the natural language text;
filling slot position information by adopting a slot position type corresponding to the target service type;
and generating an endowment service order according to the filled slot position information, and executing the endowment service order.
Storage medium-any of various types of memory electronics or storage electronics. The term "storage medium" is intended to include: mounting media such as CD-ROM, floppy disk, or tape devices; computer system memory or random access memory such as DRAM, DDR RAM, SRAM, EDO RAM, Lanbas (Rambus) RAM, etc.; non-volatile memory such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the computer system in which the program is executed, or may be located in a different second computer system connected to the computer system through a network (such as the internet). The second computer system may provide the program instructions to the computer for execution. The term "storage medium" may include two or more storage media that may reside in different unknowns (e.g., in different computer systems connected by a network). The storage medium may store program instructions (e.g., embodied as a computer program) that are executable by one or more processors.
Of course, the storage medium provided in the embodiments of the present application contains computer-executable instructions, and the computer-executable instructions are not limited to the above-mentioned human-computer interaction operations of the endowment service, and may also perform related operations in the human-computer interaction method of the endowment service provided in any embodiments of the present application.
The embodiment of the application provides electronic equipment, and a man-machine conversation device of the endowment service provided by the embodiment of the application can be integrated in the electronic equipment. Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 5, the present embodiment provides an electronic device 500, which includes: one or more processors 520; the storage device 510 is configured to store one or more programs, and when the one or more programs are executed by the one or more processors 520, the one or more processors 520 implement a man-machine conversation method of the endowment service provided by the embodiment of the present application, the method includes:
receiving voice content of a user, and converting the voice content into a natural language text;
determining a target service type according to the natural language text;
filling slot position information by adopting a slot position type corresponding to the target service type;
and generating an endowment service order according to the filled slot position information, and executing the endowment service order.
Of course, those skilled in the art will understand that the processor 520 also implements the technical solution of the man-machine conversation method of the endowment service provided by any embodiment of the present application.
The electronic device 500 shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 5, the electronic device 500 includes a processor 520, a storage 510, an input 530, and an output 540; the number of the processors 520 in the electronic device may be one or more, and one processor 520 is taken as an example in fig. 5; the processor 520, the storage 510, the input device 530, and the output device 540 in the electronic apparatus may be connected by a bus or other means, and are exemplified by a bus 550 in fig. 5.
The storage device 510 is a computer-readable storage medium, and can be used to store software programs, computer-executable programs, and module units, such as program instructions corresponding to the man-machine conversation method of the endowment service in the embodiment of the present application.
The storage device 510 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the storage 510 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, storage 510 may further include memory located remotely from processor 520, which may be connected via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 530 may be used to receive input numbers, character information, or voice information, and to generate key signal inputs related to user settings and function control of the electronic apparatus. The output device 540 may include a display screen, speakers, etc. of electronic equipment.
The electronic equipment that this application embodiment provided can be through man-machine conversation's means, and is quick discernment and processing to endowment service demand to reach the improvement and serve demand treatment effeciency to endowment, and reduce the purpose of the human cost of endowment service.
The man-machine conversation device, the storage medium and the electronic equipment of the endowment service provided by the embodiments can execute the man-machine conversation method of the endowment service provided by any embodiment of the application, and have corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in the above embodiments may be referred to a man-machine conversation method of the endowment service provided in any embodiments of the present application.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present application and the technical principles employed. It will be understood by those skilled in the art that the present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the application. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present application is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present application, and the scope of the present application is determined by the scope of the appended claims.
Claims (15)
1. A method for human-computer conversation for an aging service, the method comprising:
receiving voice content of a user, and converting the voice content into a natural language text;
determining a target service type according to the natural language text;
filling slot position information by adopting a slot position type corresponding to the target service type;
and generating an endowment service order according to the filled slot position information, and executing the endowment service order.
2. The method of claim 1, wherein after determining the target service type, the method further comprises:
determining a user intention from the target service type according to the natural language text;
correspondingly, generating an endowment service order according to the filled slot position information, and executing the endowment service order, wherein the endowment service order comprises the following steps:
and generating an endowment service order according to the filled slot position information and the user intention, and executing the endowment service order.
3. The method of claim 1, wherein receiving speech content of a user, and converting the speech content into natural language text comprises:
performing effective information extraction pretreatment on the voice content to obtain effective voice content with silence at two ends removed;
performing framing processing on the effective voice content to obtain an effective voice frame;
and processing the effective voice frame to obtain a natural language text.
4. The method of claim 3, wherein processing the valid speech frames to obtain natural language text comprises:
performing characteristic extraction on the effective speech frame through a linear prediction cepstrum coefficient;
and decoding the extracted features by adopting an acoustic model and/or a language model to obtain a natural language text.
5. The method of claim 4, wherein the acoustic model comprises a hidden Markov model; the language model includes an n-gram model.
6. The method of claim 1, wherein the target service type comprises: at least one of an endowment service order, an endowment service evaluation, a community endowment campaign, and an endowment service recommendation.
7. The method of claim 6, wherein determining a target service type from the natural language text comprises:
extracting corpus feature vectors of the natural language text by adopting a TF-IDF algorithm;
and inputting the corpus feature vector into a support vector machine, and determining a target service type according to an output result of the support vector machine.
8. The method of claim 1, wherein populating slot information with a slot type corresponding to the target service type comprises:
and filling slot position information by adopting a recurrent neural network model according to the natural language text.
9. The method of claim 2, wherein after filling slot information with a slot type corresponding to the target service type and after determining a user intent from the target service type from the natural language text, the method further comprises:
and determining to execute an action according to the slot position information and the user intention.
10. The method of claim 9, wherein the performing act comprises: executing tasks, clarifying slot positions, confirming slot values and failing matching.
11. The method of claim 10, wherein if the performing act is performing a service, then performing:
and generating an endowment service order according to the filled slot position information and the user intention, and executing the endowment service order.
12. The method of claim 10, wherein if the execution action is to clarify the slot, issue a question statement according to the slot value information of the slot;
and if the execution action is to confirm the slot value, sending the information of the slot value to be confirmed to the user for the user to confirm.
13. A human-machine conversation apparatus for a endowment service, the apparatus comprising:
the natural language text conversion module is used for receiving the voice content of a user and converting the voice content into a natural language text;
the target service type determining module is used for determining a target service type according to the natural language text;
the slot position information filling module is used for filling the slot position information by adopting a slot position type corresponding to the target service type;
and the order generation module is used for generating an endowment service order according to the filled slot position information and executing the endowment service order.
14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a man-machine conversation method of an endowment service according to any one of claims 1 to 12.
15. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements a man-machine conversation method of an endowment service according to any one of claims 1 to 12 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011034916.0A CN112199498A (en) | 2020-09-27 | 2020-09-27 | Man-machine conversation method, device, medium and electronic equipment for endowment service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011034916.0A CN112199498A (en) | 2020-09-27 | 2020-09-27 | Man-machine conversation method, device, medium and electronic equipment for endowment service |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112199498A true CN112199498A (en) | 2021-01-08 |
Family
ID=74007384
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011034916.0A Pending CN112199498A (en) | 2020-09-27 | 2020-09-27 | Man-machine conversation method, device, medium and electronic equipment for endowment service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112199498A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112929502A (en) * | 2021-02-05 | 2021-06-08 | 国家电网有限公司客户服务中心 | Voice recognition method and system based on electric power customer service |
| CN114338921A (en) * | 2021-12-31 | 2022-04-12 | 济南丽阳神州智能科技有限公司 | Home service-based incoming call processing method, device and medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107248119A (en) * | 2017-04-28 | 2017-10-13 | 深圳市前海安测信息技术有限公司 | Endowment service processing system and method based on Internet of Things |
| CN107436875A (en) * | 2016-05-25 | 2017-12-05 | 华为技术有限公司 | File classification method and device |
| CN109597607A (en) * | 2018-10-31 | 2019-04-09 | 拓科(武汉)智能技术股份有限公司 | Task interactive system and its implementation, device and electronic equipment |
| CN110211001A (en) * | 2019-05-17 | 2019-09-06 | 深圳追一科技有限公司 | A kind of hotel assistant customer service system, data processing method and relevant device |
| CN110473531A (en) * | 2019-09-05 | 2019-11-19 | 腾讯科技（深圳）有限公司 | Audio recognition method, device, electronic equipment, system and storage medium |
| CN110659970A (en) * | 2018-06-12 | 2020-01-07 | 百度在线网络技术（北京）有限公司 | Account information processing method and device based on voice recognition and electronic equipment |
| CN110659361A (en) * | 2019-10-11 | 2020-01-07 | 北京物灵科技有限公司 | Conversation method, device, equipment and medium |
| CN110737762A (en) * | 2019-10-09 | 2020-01-31 | 尹曦 | old people personal information assistant system based on voice interaction |
-
2020
- 2020-09-27 CN CN202011034916.0A patent/CN112199498A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107436875A (en) * | 2016-05-25 | 2017-12-05 | 华为技术有限公司 | File classification method and device |
| CN107248119A (en) * | 2017-04-28 | 2017-10-13 | 深圳市前海安测信息技术有限公司 | Endowment service processing system and method based on Internet of Things |
| CN110659970A (en) * | 2018-06-12 | 2020-01-07 | 百度在线网络技术（北京）有限公司 | Account information processing method and device based on voice recognition and electronic equipment |
| CN109597607A (en) * | 2018-10-31 | 2019-04-09 | 拓科(武汉)智能技术股份有限公司 | Task interactive system and its implementation, device and electronic equipment |
| CN110211001A (en) * | 2019-05-17 | 2019-09-06 | 深圳追一科技有限公司 | A kind of hotel assistant customer service system, data processing method and relevant device |
| CN110473531A (en) * | 2019-09-05 | 2019-11-19 | 腾讯科技（深圳）有限公司 | Audio recognition method, device, electronic equipment, system and storage medium |
| CN110737762A (en) * | 2019-10-09 | 2020-01-31 | 尹曦 | old people personal information assistant system based on voice interaction |
| CN110659361A (en) * | 2019-10-11 | 2020-01-07 | 北京物灵科技有限公司 | Conversation method, device, equipment and medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112929502A (en) * | 2021-02-05 | 2021-06-08 | 国家电网有限公司客户服务中心 | Voice recognition method and system based on electric power customer service |
| CN114338921A (en) * | 2021-12-31 | 2022-04-12 | 济南丽阳神州智能科技有限公司 | Home service-based incoming call processing method, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200043483A1 (en) | Minimum word error rate training for attention-based sequence-to-sequence models | |
| US20180308487A1 (en) | Dialogue System Incorporating Unique Speech to Text Conversion Method for Meaningful Dialogue Response | |
| US9514126B2 (en) | Method and system for automatically detecting morphemes in a task classification system using lattices | |
| US11450332B2 (en) | Audio conversion learning device, audio conversion device, method, and program | |
| US7143035B2 (en) | Methods and apparatus for generating dialog state conditioned language models | |
| US11580145B1 (en) | Query rephrasing using encoder neural network and decoder neural network | |
| CN109903750B (en) | Voice recognition method and device | |
| CN111402891A (en) | Speech recognition method, apparatus, device and storage medium | |
| US6341264B1 (en) | Adaptation system and method for E-commerce and V-commerce applications | |
| WO2023222088A1 (en) | Voice recognition and classification method and apparatus | |
| CN114038457B (en) | Method, electronic device, storage medium, and program for voice wakeup | |
| US11961515B2 (en) | Contrastive Siamese network for semi-supervised speech recognition | |
| CN112071310A (en) | Speech recognition method and apparatus, electronic device, and storage medium | |
| CN112199498A (en) | Man-machine conversation method, device, medium and electronic equipment for endowment service | |
| CN114550703A (en) | Training method and device of voice recognition system, and voice recognition method and device | |
| CN111862943B (en) | Speech recognition method and device, electronic equipment and storage medium | |
| US11715458B2 (en) | Efficient streaming non-recurrent on-device end-to-end model | |
| Rose et al. | Integration of utterance verification with statistical language modeling and spoken language understanding | |
| CN111400463A (en) | Dialog response method, apparatus, device and medium | |
| WO2023222090A1 (en) | Information pushing method and apparatus based on deep learning | |
| CN111883133B (en) | Customer service voice recognition method, customer service voice recognition device, server and storage medium | |
| US20240144921A1 (en) | Domain specific neural sentence generator for multi-domain virtual assistants | |
| CN112447169B (en) | Word boundary estimation method and device and electronic equipment | |
| US20230107695A1 (en) | Fusion of Acoustic and Text Representations in RNN-T | |
| CN113555016A (en) | Voice interaction method, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |