CN114155853A - Rejection method, device, equipment and storage medium - Google Patents
Rejection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114155853A CN114155853A CN202111494300.6A CN202111494300A CN114155853A CN 114155853 A CN114155853 A CN 114155853A CN 202111494300 A CN202111494300 A CN 202111494300A CN 114155853 A CN114155853 A CN 114155853A
- Authority
- CN
- China
- Prior art keywords
- target
- target information
- information
- intention
- intention type
- 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 52
- 238000013145 classification model Methods 0.000 claims abstract description 24
- 230000006870 function Effects 0.000 claims description 75
- 238000012549 training Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 8
- 230000003993 interaction Effects 0.000 abstract description 35
- 230000000694 effects Effects 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 230000000977 initiatory effect Effects 0.000 description 6
- 238000003062 neural network model Methods 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000002452 interceptive effect Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/1815—Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
-
- 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
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/1822—Parsing for meaning understanding
-
- 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)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a rejection method, a rejection device, rejection equipment and a storage medium, wherein the method comprises the following steps: acquiring target information to be identified; inputting the target information into a pre-trained function recognition model to obtain a target service function corresponding to the target information; inputting the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information; and determining whether to reject the target information according to the target service function and the target intention type. According to the rejection method provided by the invention, whether the target information is in a functional range or a non-functional range according to the target service type is distinguished by judging whether the target intention type corresponding to the target information meets the intention type supported by the target service function corresponding to the target information, and the rejection is carried out on the non-functional range, so that the accuracy of the voice interaction system for rejecting the information is improved, the rejection task in the voice interaction system is optimized, and the rejection effect is improved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a rejection method, a rejection device, rejection equipment and a storage medium.
Background
With the progress of artificial intelligence technology, man-machine voice interaction has also been developed, in order to improve the user experience, full-duplex voice interaction mode begins to appear gradually, and the full-duplex voice interaction mode is characterized in that after one-time awakening, multiple continuous interactions can be carried out within a certain time.
In the full-duplex voice interaction mode, the intelligent voice system needs to continuously input the voice input by the user, and perform natural language understanding and replying on the recorded voice. In this mode, it is necessary to accurately recognize some voices input by the user and not interacting with the system (e.g., chatting with other people, self-speaking and self-speaking), and reject the voice of the user when the user is determined not to interact with the system, so as to avoid disturbing the user.
To implement the above functionality, current intelligent speech systems typically distinguish between functional-Domain speech and non-functional-Domain speech (i.e., Out-of-function speech) by building classification models, such as Out-of-Domain (OOD) models. Such classification models are mainly based on the language content, for example, it can be distinguished whether a user wants to listen to a song or ask weather, or whether a sentence and function relate to the content completely, but when aiming at similar information content and different intentions, namely the language materials (for example, "i want to go to the Shanghai" and "i go to the Shanghai"), the distinguishing effect is often not good, so that the recognition rejection effect is poor, and the user experience is affected.
Disclosure of Invention
In view of the foregoing problems in the prior art, an object of the present invention is to provide a method, an apparatus, a device, and a storage medium for rejecting information, which can improve the accuracy of rejecting information, and further improve the interaction effect between a user and a voice interaction system.
In order to solve the above problem, the present invention provides a rejection method, including:
acquiring target information to be identified;
inputting the target information into a pre-trained function recognition model to obtain a target service function corresponding to the target information;
inputting the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information;
and determining whether to reject the target information according to the target service function and the target intention type.
Further, the acquiring target information to be identified includes:
acquiring voice information to be recognized;
and carrying out voice recognition processing on the voice information to obtain a voice text corresponding to the voice information.
Further, the determining whether to reject the target information according to the target business function and the target intention type includes:
acquiring an intention type set corresponding to the target service function;
matching the target intention type with the intention type set to obtain a matching result;
and determining whether the target information is rejected or not according to the matching result.
Further, the matching the target intention type with the intention type set to obtain a matching result includes:
matching the target intention type with each intention type in the intention type set respectively;
when the target intention type is matched with any intention type in the intention type set, determining that the matching result is successful;
and when the target intention type is not matched with each intention type in the intention type set, determining that the matching result is matching failure.
Further, the determining whether to reject the target information according to the matching result includes:
refusing to identify the target information when the matching result is matching failure;
and when the matching result is that the matching is successful, obtaining and outputting semantic information corresponding to the target information.
Further, the method further comprises:
and training the function recognition model and the intention classification model by using a sample acquired in advance.
Another aspect of the present invention provides a rejection apparatus, including:
the acquisition module is used for acquiring target information to be identified;
the identification module is used for inputting the target information into a pre-trained function identification model to obtain a target service function corresponding to the target information;
the classification module is used for inputting the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information;
and the determining module is used for determining whether to reject the target information according to the target service function and the target intention type.
Further, the apparatus further comprises:
and the training module is used for training the function recognition model and the intention classification model by using samples acquired in advance.
Another aspect of the present invention provides an electronic device, including a processor and a memory, where at least one instruction or at least one program is stored in the memory, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the rejection method as described above.
Another aspect of the present invention provides a computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the rejection method as described above.
Due to the technical scheme, the invention has the following beneficial effects:
according to the rejection method provided by the embodiment of the invention, the target service function corresponding to the target information to be recognized is determined by using the function recognition model, the target intention type corresponding to the target information to be recognized is determined by using the intention classification model, whether the target information is in the functional range of the target service type or in the non-functional range of the target service type is distinguished by judging whether the target intention type is in accordance with the intention type supported by the target service function, finally rejection is carried out on the non-functional range of the target service type, a better distinguishing effect is achieved on the telephone term materials with similar information content and different intentions, the accuracy of the speech interaction system in rejecting information is improved, the rejection task in the speech interaction system is optimized, the rejection effect is improved, further unnecessary disturbance on a user can be avoided, and the interaction effect between the user and the speech interaction system is improved, the user experience is improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings used in the description of the embodiment or the prior art will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is a schematic illustration of an implementation environment provided by an embodiment of the invention;
FIG. 2 is a flow chart of a rejection method provided by one embodiment of the invention;
FIG. 3 is a flow chart of a rejection method provided by another embodiment of the invention;
fig. 4 is a schematic structural diagram of a rejection apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural view of a rejection apparatus according to another embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or device that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or device.
In order to make the objects, technical solutions and advantages disclosed in the embodiments of the present invention more clearly apparent, the embodiments of the present invention are described in further detail below with reference to the accompanying drawings and the embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the invention and are not intended to limit the embodiments of the invention. First, the embodiments of the present invention explain the following concepts:
full duplex interaction: after the voice system is awakened once, the user can carry out continuous interaction for many times without awakening the voice system within a certain time.
Natural Language Understanding (NLU): the NLU is an application field of artificial intelligence, and is a technology for communicating with a computer by using natural language, and the computer can 'understand' the natural language through the NLU to further execute some language functions expected by human beings, in other words, the NLU is a bridge for people to communicate with machines. NLUs may include, but are not limited to, the following: firstly, the correct order rule and concept of sentences can be understood, and sentences without rules can also be understood; knowing the exact meaning, form, part of speech and word-forming method of the word; thirdly, understanding semantic classification of words, ambiguity of words and ambiguity of words; (iv) specified and indefinite properties and all properties; structure knowledge and practical concepts in the problem domain; sixthly, the tone information and rhythm expression of the language; seventhly, character knowledge related to language expression forms; the background knowledge of the discourse domain.
Speech intent (speech acts) classification theory: the pragmatics theory focuses on behaviors carried by the speech except information, and performs intent level classification on the behaviors.
Referring to the specification, fig. 1 is a schematic diagram illustrating an implementation environment provided by an embodiment of the present invention, and as shown in fig. 1, the implementation environment may include at least one terminal device 110 and a server 120. The server 120 and each terminal device 110 may be directly or indirectly connected through wired or wireless communication, which is not limited in this embodiment of the present invention.
The terminal device 110 may include a smart phone, a tablet computer, a notebook computer, a desktop computer, a digital assistant, a smart speaker, a smart wearable device, a vehicle-mounted terminal, a server, and other types of physical devices, and may also include software running in the physical devices, such as an application program, but is not limited thereto. The operating system running on the terminal device 110 may include, but is not limited to, an android system, an IOS system, a linux system, a windows system, and the like.
The server 120 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.
In practical application, the rejection method provided by the embodiment of the invention can be applied to a full-duplex interaction scene. Specifically, the terminal device 110 may be configured with an intelligent voice system, the intelligent voice system may acquire voice information sent by a user through a voice acquisition module, and send the acquired voice information to the server 120 for processing, the server 120 may perform rejection judgment on the voice information through the rejection method provided in the embodiment of the present invention, determine whether the voice information is voice interacting with the intelligent voice system, reject recognition for voice not interacting with the intelligent voice system, give no feedback, output corresponding semantic information for voice interacting with the intelligent voice system, and further generate corresponding interaction data to send to the terminal device 110, so that the intelligent voice system completes an interaction behavior with the user.
It should be noted that fig. 1 is only an example. Those skilled in the art will appreciate that although only 2 terminal devices 110 are shown in fig. 1, this is not a limitation of the embodiments of the present invention and that more or fewer terminal devices 110 may be included than shown.
Referring to the specification, fig. 2 illustrates a flow of a rejection method according to an embodiment of the present invention, which may be applied to the server 120 in fig. 1, and specifically as shown in fig. 2, the method may include the following steps:
s210: and acquiring target information to be identified.
In the embodiment of the invention, the target information is information which needs to be judged whether to reject identification, and the target information can be various information input by a user in the interaction process of the intelligent voice system and the user.
Specifically, the acquiring target information to be identified may include:
acquiring voice information to be recognized;
and carrying out voice recognition processing on the voice information to obtain a voice text corresponding to the voice information.
Specifically, terminal equipment can be configured with intelligent voice system, intelligent voice system can gather user input's speech information and send to the server through pronunciation collection module, and the server can adopt speech recognition technology right speech information carries out identification processing, obtains the speech text that speech information corresponds, and will the speech text is as the target information of treating discernment. The voice acquisition module can be a sound sensor, a microphone and the like.
S220: and inputting the target information into a pre-trained function recognition model to obtain a target service function corresponding to the target information.
In the embodiment of the present invention, the function identification model may process input target information, determine an intention corresponding to the target information, and then determine a target service function corresponding to the target information according to the intention corresponding to the target information. The target service function may be one or more of various functions provided by the intelligent voice system to the user, and various functions provided by the intelligent voice system to the user may include various functions common in the prior art, such as a weather query function, a music playing function, a navigation initiating function, and the like, which are not described herein again in the embodiments of the present invention.
In one possible embodiment, the method may further include the step of training the function recognition model using pre-acquired samples.
Specifically, first training data labeled with a corresponding business function may be obtained in advance, and the first training data may be used to train a first preset neural network model to obtain the function recognition model. The first preset neural network model may include, but is not limited to, a deep neural network model commonly used in the prior art, and the details of the embodiment of the present invention are not repeated herein.
S230: and inputting the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information.
In the embodiment of the present invention, the intention classification model may process input target information, and determine a target intention type corresponding to the target information. The target intention type can be one or more of various intention types in a classification framework under the current application scene, and the classification framework under the current application scene can analyze user speech information under the current application scene to be predetermined based on a speech intention classification theory.
For an automobile scene, the user speech information in the vehicle-mounted intelligent voice system can comprise a speech technology for initiating an instruction and other non-instruction speech technologies, the classification frame under the automobile scene can be determined to comprise a speech technology for initiating an instruction, a speech technology for willingness expression, an information question and the like according to the speech technology for initiating the instruction, and the classification frame under the automobile scene can be determined to further comprise an event statement class, an emotion expression class and the like according to the other non-instruction speech technologies.
It should be noted that the classification frameworks in different application scenarios may include the same intent type or may include different intent types, which is not limited in this embodiment of the present invention.
In one possible embodiment, the method may further include training the intent classification model using pre-acquired samples.
Specifically, second training data labeled with a corresponding intention type may be obtained in advance, and the second training data may be used to train a second preset neural network model to obtain the intention classification model. The second preset neural network model may include, but is not limited to, a deep neural network model commonly used in the prior art, and the details of the embodiment of the present invention are not repeated herein.
In the embodiment of the invention, the intention classification model has a better distinguishing effect on similar information content and dialectical information with different intentions. For example, the same event "listen to music" may have different expression of the types of intentions, as shown in table 1, and it is the task of the intent classification model to distinguish these types of intentions.
TABLE 1
| Phonetics information | Type of intention | Whether to correspond to a business function |
| Listening to music | Initiating command (command) | Corresponding music playing function |
| I want to listen to music | Expressing will (desire) | Corresponding music playing function |
| I do not need to listen to music | Expressing will (desire) | Corresponding to the function of pausing playing music |
| I love listening to music | Expression preference (preference) | Non-corresponding business function |
| I listen to music frequently | Stating something (info) | Non-corresponding business function |
| Do you listen to music | Ask others (request) | Non-corresponding business function |
Through the intention classification model, when a user expresses the same event but different intentions, whether the user intends to initiate an instruction or not (such as chatting with other people) can be distinguished more accurately, and the accurate judgment on whether the user initiates the instruction to the system can better realize the full-duplex wake-free function of the intelligent voice system.
It should be noted that, step S220 and step S230 may be sequentially executed according to the above-mentioned order, may also be sequentially executed according to another order, and may also be executed in parallel, which is not limited in this embodiment of the present invention.
S240: and determining whether to reject the target information according to the target service function and the target intention type.
Specifically, as shown in fig. 3, the determining whether to reject the target information according to the target service function and the target intention type may include:
s241: and acquiring an intention type set corresponding to the target service function.
In the embodiment of the present invention, various intention types in a classification framework in a current application scenario and corresponding logics of various service functions supported by an intelligent voice system may be predetermined, and specifically, for each service function supported by the intelligent voice system, there is a function scope terminology for the function, so that the function scope terminology corresponding to each service function may be analyzed to summarize several specific intention types, and an intention type set corresponding to the service function is obtained. Illustratively, for example, the function for querying weather includes an information query jargon "how weather is today", or a command jargon "report weather today", and the like, so that the intention type set corresponding to the function for querying weather may be determined to include an information query type and an initiating command type, and the like.
S242: and matching the target intention type with the intention type set to obtain a matching result.
In this embodiment of the present invention, the matching the target intention type with the intention type set to obtain a matching result may include:
matching the target intention type with each intention type in the intention type set respectively;
when the target intention type is matched with any intention type in the intention type set, determining that the matching result is successful;
and when the target intention type is not matched with each intention type in the intention type set, determining that the matching result is matching failure.
In this embodiment of the present invention, the target intent type may be matched with each intent type in the intent type set one by one, and it is determined whether the target intent type is the same as any one of the intent types in the intent type set, that is, whether the target intent type is included in the intent type set is determined, if the target intent type is included in the intent type set, the matching is successful, otherwise, the matching is failed.
S243: and determining whether the target information is rejected or not according to the matching result.
In this embodiment of the present invention, the determining whether to reject the target information according to the matching result may include:
refusing to identify the target information when the matching result is matching failure;
and when the matching result is that the matching is successful, obtaining and outputting semantic information corresponding to the target information.
In the embodiment of the present invention, when the target intent type is included in the intent type set, it may be determined that the target information is a functional scope for the target service type, that is, the target information needs to be identified, and at this time, semantic information corresponding to the target information may be determined based on a natural language understanding technology and output.
In the embodiment of the present invention, when the target intention type is not included in the intention type set, it may be determined that the target information is a non-functional range technology, that is, the target information does not need to be identified, and at this time, the target information may be rejected from being identified.
Exemplarily, assuming that target information input by a user is "i want to go to the people square", determining that a target business function corresponding to the target information is "initiate navigation function" through a function identification model, determining that a target intention type corresponding to the target information is "initiate command type" through an intention classification model, and assuming that an intention type set corresponding to the "initiate navigation function" includes "initiate command type", "will express type", and "information question type", a result of matching the target intention type with the intention type set is successful, that is, the target information needs to be identified and an interactive behavior needs to be initiated. Assuming that target information input by a user is 'i go to a people square' or 'i go to a people square again tomorrow', determining that a target intention type corresponding to the target information is an 'event statement type' through an intention classification model, and determining that a result of matching the target intention type with the intention type set is matching failure, namely, the target information does not need to be identified and interactive behavior is initiated.
In a possible embodiment, the method may further include generating interaction information for the target information according to the semantic information, and sending the interaction information to a terminal device, so that a voice interaction system in the terminal device performs a corresponding interaction operation according to the interaction information. The interaction information may include reply information for the target information, and may also include an interaction operation for a target execution object. When the interactive information is reply information, the voice interactive system may display the reply information, for example, display the reply information in a designated interface, or broadcast the reply information by voice. When the interaction information is an interaction operation for a target execution object, the voice interaction system may generate an operation instruction for executing the interaction operation and send the operation instruction to the target execution object, and the target execution object may execute the interaction operation in response to the operation instruction to obtain an execution result.
In summary, according to the rejection method provided by the embodiment of the present invention, the function recognition model is used to determine the target service function corresponding to the target information to be recognized, the intention classification model is used to determine the target intention type corresponding to the target information to be recognized, and whether the target intention type matches the intention type supported by the target service function is determined, so as to distinguish whether the target information is directed to the functional scope dialect or the non-functional scope dialect of the target service type, and finally reject the non-functional scope dialect, so that the term material with similar information content and different intentions has a better distinguishing effect, the accuracy of rejecting the information by the voice interaction system is improved, the rejection task in the voice interaction system is optimized, the rejection effect is improved, and further unnecessary disturbance to the user can be avoided, the interaction effect between the user and the voice interaction system is improved, and the user experience is improved.
Referring to the specification and to fig. 4, there is shown a structure of a rejection apparatus 400 according to an embodiment of the present invention. As shown in fig. 4, the apparatus 400 may include:
an obtaining module 410, configured to obtain target information to be identified;
the recognition module 420 is configured to input the target information into a pre-trained function recognition model to obtain a target service function corresponding to the target information;
the classification module 430 is configured to input the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information;
a determining module 440, configured to determine whether to reject the target information according to the target service function and the target intention type.
In one possible embodiment, as shown in fig. 5, the apparatus 400 may further include:
a training module 450, configured to train the function recognition model and the intention classification model using samples acquired in advance.
It should be noted that, when the apparatus provided in the foregoing embodiment implements the functions thereof, only the division of the functional modules is illustrated, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the internal structure of the apparatus may be divided into different functional modules to implement all or part of the functions described above. In addition, the apparatus provided in the above embodiments and the corresponding method embodiments belong to the same concept, and specific implementation processes thereof are detailed in the corresponding method embodiments and are not described herein again.
An embodiment of the present invention further provides an electronic device, which includes a processor and a memory, where the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the rejection method provided by the above method embodiment.
The memory may be used to store software programs and modules, and the processor may execute various functional applications and data processing by operating the software programs and modules stored in the memory. The memory can mainly comprise a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs needed by functions and the like; the storage data area may store data created according to use of the apparatus, and the like. Further, the memory 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 volatile solid state storage device. Accordingly, the memory may also include a memory controller to provide the processor access to the memory.
The method embodiments provided by the embodiments of the present invention may be executed in a terminal, a server, or a similar computing device, that is, the electronic device may include a terminal, a server, or a similar computing device. Taking the operation on the server as an example, as shown in fig. 6, it shows a schematic structural diagram of the server implementing the rejection method provided by the embodiment of the present invention. The server 600 may vary significantly due to configuration or performance, and may include one or more Central Processing Units (CPUs) 610 (e.g., one or more processors) and memory 630, one or more storage media 620 (e.g., one or more mass storage devices) storing applications 623 or data 622. Memory 630 and storage medium 620 may be, among other things, transient or persistent storage. The program stored on the storage medium 620 may include one or more modules, each of which may include a series of instruction operations for the server. Still further, the central processor 610 may be configured to communicate with the storage medium 620 to execute a series of instruction operations in the storage medium 620 on the server 600. The server 600 may also include one or more power supplies 660, one or more wired or wireless network interfaces 650, one or more input-output interfaces 640, and/or one or more operating systems 621, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and so forth.
The input/output interface 640 may be used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the server 600. In one example, i/o Interface 640 includes a Network adapter (NIC) that may be coupled to other Network devices via a base station to communicate with the internet. In one example, the i/o interface 640 may be a Radio Frequency (RF) module for communicating with the internet in a wireless manner, and the wireless communication may use any communication standard or protocol, including but not limited to Global System for Mobile communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Message Service (SMS), and the like.
Those skilled in the art will appreciate that the configuration shown in fig. 6 is merely illustrative and that server 600 may include more or fewer components than shown in fig. 6 or have a different configuration than shown in fig. 6.
An embodiment of the present invention further provides a computer-readable storage medium, which may be disposed in an electronic device to store at least one instruction or at least one program for implementing a rejection method, where the at least one instruction or the at least one program is loaded and executed by the processor to implement the rejection method provided by the foregoing method embodiment.
Optionally, in an embodiment of the present invention, the storage medium may include, but is not limited to: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
An embodiment of the invention also provides a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the rejection method provided in the various alternative embodiments described above.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A method of rejecting comprising:
acquiring target information to be identified;
inputting the target information into a pre-trained function recognition model to obtain a target service function corresponding to the target information;
inputting the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information;
and determining whether to reject the target information according to the target service function and the target intention type.
2. The method of claim 1, wherein the obtaining target information to be identified comprises:
acquiring voice information to be recognized;
and carrying out voice recognition processing on the voice information to obtain a voice text corresponding to the voice information.
3. The method of claim 1, wherein the determining whether to reject the target information according to the target business function and the target intent type comprises:
acquiring an intention type set corresponding to the target service function;
matching the target intention type with the intention type set to obtain a matching result;
and determining whether the target information is rejected or not according to the matching result.
4. The method of claim 3, wherein matching the target intent type with the set of intent types, resulting in a matching result comprises:
matching the target intention type with each intention type in the intention type set respectively;
when the target intention type is matched with any intention type in the intention type set, determining that the matching result is successful;
and when the target intention type is not matched with each intention type in the intention type set, determining that the matching result is matching failure.
5. The method of claim 3, wherein the determining whether to reject the target information according to the matching result comprises:
refusing to identify the target information when the matching result is matching failure;
and when the matching result is that the matching is successful, obtaining and outputting semantic information corresponding to the target information.
6. The method of claim 1, further comprising:
and training the function recognition model and the intention classification model by using a sample acquired in advance.
7. A rejection device, comprising:
the acquisition module is used for acquiring target information to be identified;
the identification module is used for inputting the target information into a pre-trained function identification model to obtain a target service function corresponding to the target information;
the classification module is used for inputting the target information into a pre-trained intention classification model to obtain a target intention type corresponding to the target information;
and the determining module is used for determining whether to reject the target information according to the target service function and the target intention type.
8. The apparatus of claim 7, further comprising:
and the training module is used for training the function recognition model and the intention classification model by using samples acquired in advance.
9. An electronic device, comprising a processor and a memory, wherein at least one instruction or at least one program is stored in the memory, and the at least one instruction or the at least one program is loaded by the processor and executed to implement the denial method according to any of claims 1-6.
10. A computer-readable storage medium, wherein at least one instruction or at least one program is stored in the computer-readable storage medium, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the denial method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111494300.6A CN114155853A (en) | 2021-12-08 | 2021-12-08 | Rejection method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111494300.6A CN114155853A (en) | 2021-12-08 | 2021-12-08 | Rejection method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114155853A true CN114155853A (en) | 2022-03-08 |
Family
ID=80453495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111494300.6A Pending CN114155853A (en) | 2021-12-08 | 2021-12-08 | Rejection method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114155853A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115376513A (en) * | 2022-10-19 | 2022-11-22 | 广州小鹏汽车科技有限公司 | Voice interaction method, server and computer readable storage medium |
| CN116959421A (en) * | 2023-09-21 | 2023-10-27 | 湖北星纪魅族集团有限公司 | Method and device for processing audio data, audio data processing equipment and medium |
| WO2024078460A1 (en) * | 2022-10-13 | 2024-04-18 | 广州小鹏汽车科技有限公司 | Speech processing method, speech interaction method, server, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109871446A (en) * | 2019-01-31 | 2019-06-11 | 平安科技（深圳）有限公司 | Rejection method for identifying, electronic device and storage medium in intention assessment |
| CN111625636A (en) * | 2020-05-28 | 2020-09-04 | 深圳追一科技有限公司 | Man-machine conversation refusal identification method, device, equipment and medium |
| CN111884887A (en) * | 2020-07-01 | 2020-11-03 | 海尔优家智能科技（北京）有限公司 | Voice interaction method and device, storage medium and electronic device |
| CN112667076A (en) * | 2020-12-23 | 2021-04-16 | 广州橙行智动汽车科技有限公司 | Voice interaction data processing method and device |
| US20210303263A1 (en) * | 2020-03-26 | 2021-09-30 | Hyundai Motor Company | Dialogue system and vehicle having the same, and method of controlling dialogue system |
-
2021
- 2021-12-08 CN CN202111494300.6A patent/CN114155853A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109871446A (en) * | 2019-01-31 | 2019-06-11 | 平安科技（深圳）有限公司 | Rejection method for identifying, electronic device and storage medium in intention assessment |
| WO2020155766A1 (en) * | 2019-01-31 | 2020-08-06 | 平安科技（深圳）有限公司 | Method, device and apparatus for identification rejection in intention identification, and storage medium |
| US20210303263A1 (en) * | 2020-03-26 | 2021-09-30 | Hyundai Motor Company | Dialogue system and vehicle having the same, and method of controlling dialogue system |
| CN111625636A (en) * | 2020-05-28 | 2020-09-04 | 深圳追一科技有限公司 | Man-machine conversation refusal identification method, device, equipment and medium |
| CN111884887A (en) * | 2020-07-01 | 2020-11-03 | 海尔优家智能科技（北京）有限公司 | Voice interaction method and device, storage medium and electronic device |
| CN112667076A (en) * | 2020-12-23 | 2021-04-16 | 广州橙行智动汽车科技有限公司 | Voice interaction data processing method and device |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024078460A1 (en) * | 2022-10-13 | 2024-04-18 | 广州小鹏汽车科技有限公司 | Speech processing method, speech interaction method, server, and storage medium |
| CN115376513A (en) * | 2022-10-19 | 2022-11-22 | 广州小鹏汽车科技有限公司 | Voice interaction method, server and computer readable storage medium |
| CN116959421A (en) * | 2023-09-21 | 2023-10-27 | 湖北星纪魅族集团有限公司 | Method and device for processing audio data, audio data processing equipment and medium |
| CN116959421B (en) * | 2023-09-21 | 2023-12-19 | 湖北星纪魅族集团有限公司 | Method and device for processing audio data, audio data processing equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10832686B2 (en) | Method and apparatus for pushing information | |
| US10404629B2 (en) | Automatic reply method, device, apparatus, and storage medium | |
| CN107895578B (en) | Voice interaction method and device | |
| CN114155853A (en) | Rejection method, device, equipment and storage medium | |
| KR102112814B1 (en) | Parameter collection and automatic dialog generation in dialog systems | |
| CN105895103B (en) | Voice recognition method and device | |
| CN109065052B (en) | Voice robot | |
| CN106558307A (en) | Intelligent dialogue processing equipment, method and system | |
| CN109243468B (en) | Voice recognition method and device, electronic equipment and storage medium | |
| CN110634472B (en) | Speech recognition method, server and computer readable storage medium | |
| CN109961786B (en) | Product recommendation method, device, equipment and storage medium based on voice analysis | |
| CN108737324B (en) | Method and device for generating artificial intelligence service assembly and related equipment and system | |
| KR20210128039A (en) | Speaker diarization | |
| CN111428010A (en) | Man-machine intelligent question and answer method and device | |
| CN112667796B (en) | Dialogue reply method and device, electronic equipment and readable storage medium | |
| CN111737987A (en) | Intention recognition method, device, equipment and storage medium | |
| CN109086455B (en) | Method for constructing voice recognition library and learning equipment | |
| CN110457454A (en) | A kind of dialogue method, server, conversational system and storage medium | |
| CN109545203A (en) | Audio recognition method, device, equipment and storage medium | |
| CN112364622A (en) | Dialog text analysis method, dialog text analysis device, electronic device and storage medium | |
| CN113901837A (en) | Intention understanding method, device, equipment and storage medium | |
| CN112767916A (en) | Voice interaction method, device, equipment, medium and product of intelligent voice equipment | |
| CN110931002B (en) | Man-machine interaction method, device, computer equipment and storage medium | |
| CN116821290A (en) | Multitasking dialogue-oriented large language model training method and interaction method | |
| CN110750626B (en) | Scene-based task-driven multi-turn dialogue method and system |
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 |