CN113705186B - Automatic reply method and device under message semantic analysis - Google Patents
Automatic reply method and device under message semantic analysis Download PDFInfo
- Publication number
- CN113705186B CN113705186B CN202110831527.9A CN202110831527A CN113705186B CN 113705186 B CN113705186 B CN 113705186B CN 202110831527 A CN202110831527 A CN 202110831527A CN 113705186 B CN113705186 B CN 113705186B
- Authority
- CN
- China
- Prior art keywords
- client
- category
- voice
- template
- text information
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 230000006854 communication Effects 0.000 claims abstract description 54
- 238000004891 communication Methods 0.000 claims abstract description 19
- 230000004044 response Effects 0.000 claims abstract description 11
- 239000013598 vector Substances 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 15
- 238000013145 classification model Methods 0.000 claims description 13
- 238000010801 machine learning Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008451 emotion Effects 0.000 description 6
- 230000036651 mood Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 239000003795 chemical substances by application Substances 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/186—Templates
-
- 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
-
- 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
- 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
-
- 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
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
- H04M3/5166—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing in combination with interactive voice response systems or voice portals, e.g. as front-ends
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Child & Adolescent Psychology (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Machine Translation (AREA)
Abstract
The application discloses an automatic replying method and device under message semantic analysis, wherein the method comprises the following steps: the method comprises the steps of obtaining a voice message of a client, wherein the voice message of the client is a voice file which is left by automatic response after the client performs voice communication; converting the voice message into text information; classifying the text information according to the content of the text information to obtain a first category corresponding to the text information; determining a speaking template corresponding to the first category according to the first category, wherein the speaking comprises at least one sentence; and communicating with the client according to the contact way of the client, and playing the speaking template in the communication process. The problem that the cost is high and the timeliness is poor due to the fact that a customer message needs to be replied by a manual agent when the customer message is idle in the prior art is solved, so that the cost is reduced, the response speed to the customer demand can be improved to a certain extent, and the customer experience is improved.
Description
Technical Field
The application relates to the field of language analysis, in particular to an automatic reply method and device under message semantic analysis.
Background
With the popularization of the automatic answering machine, many merchants are now automatic answering machines to answer the telephone of the customer, and if the problem of the customer can be solved according to the setting flow of the automatic answering machine, the labor cost can be saved. If the problem posed by the customer cannot be solved by the automatic answering machine setting process, the process is generally switched to manual processing.
If the transfer is performed manually, if the manual seat is not busy, the requirements of the clients can be met quickly; if the manual seat is busy, the client can wait for the message or ask the client to leave a message later. The client message is also carried out by adopting a manual reply mode, and when the manual seat is idle, the manual answer to the client question is allocated. The processing mode has high cost on one hand and poor timeliness of answering the client questions on the other hand, and reduces the experience of the client.
Disclosure of Invention
The embodiment of the application provides an automatic replying method and device under message semantic analysis, which at least solve the problems of high cost and poor timeliness caused by the fact that a customer message needs to be replied by a manual agent when the customer message is idle in the prior art.
According to one aspect of the present application, there is provided an automatic reply method under message semantic analysis, including: obtaining a voice message of a client, wherein the voice message of the client is a voice file which is left by automatic response after voice communication is carried out on the client; converting the voice message into text information; classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are preconfigured, and the first category of the text information is at least one of the categories; determining a speaking template corresponding to the first category according to the first category, wherein the speaking comprises at least one sentence; and communicating with the client according to the contact mode of the client, and playing the conversation template in the communication process.
Further, where the first category includes a plurality of categories, determining, from the first category, a conversation template corresponding to the first category includes: obtaining a conversation template corresponding to each of a plurality of categories included in the first category; and sequencing the obtained conversation templates corresponding to each category to obtain the order of playing the conversation templates by the automatic answering machine.
Further, playing the conversation template during the communication includes: acquiring introduction voice of a speaking template corresponding to each category, wherein the introduction voice is used for introducing the content of the speaking template; playing the introduction voice of the conversation template corresponding to each category to the client; receiving a conversation template selected by the customer; the conversation template selected by the client is played first during the communication.
Further, after first playing the conversation template selected by the client during the communication, the method further comprises: judging whether the client finishes the communication process, and playing other voice templates except the voice template selected by the client according to the sequence under the condition that the client does not finish the communication process.
Further, after the conversation template is played in the communication process, the method further comprises: judging whether the client finishes the communication process, if not, converting the client into an artificial seat, and adjusting the client priority to be the highest.
According to one aspect of the present application, there is provided an automatic reply device under semantic analysis of a message, including: the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring a voice message of a client, wherein the voice message of the client is a voice file which is remained through automatic response after the client performs voice communication; the conversion module is used for converting the voice message into text information; the classification module is used for classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are pre-configured, and the first category of the text information is at least one of the categories; a determining module, configured to determine a speech template corresponding to the first category according to the first category, where the speech includes at least one sentence; and the playing module is used for communicating with the client according to the contact way of the client and playing the conversation template in the communication process.
Further, the determining module is configured to: obtaining a conversation template corresponding to each of a plurality of categories included in the first category; and sequencing the obtained conversation templates corresponding to each category to obtain the order of playing the conversation templates by the automatic answering machine.
Further, the playing module is configured to: acquiring introduction voice of a speaking template corresponding to each category, wherein the introduction voice is used for introducing the content of the speaking template; playing the introduction voice of the conversation template corresponding to each category to the client; receiving a conversation template selected by the customer; the conversation template selected by the client is played first during the communication.
Further, the playing module is further configured to determine whether the client ends the communication process, and play other voice templates except the voice template selected by the client according to the sequence when the client does not end the communication process.
Further, the method further comprises the following steps: and the switching module is used for judging whether the client finishes the communication process after the conversation template is played in the communication process, if not, switching the client into an artificial seat and adjusting the priority of the client to be the highest.
In the embodiment of the application, the method comprises the steps of obtaining a voice message of a client, wherein the voice message of the client is a voice file which is remained through automatic response after voice communication is carried out by the client; converting the voice message into text information; classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are preconfigured, and the first category of the text information is at least one of the categories; determining a speaking template corresponding to the first category according to the first category, wherein the speaking comprises at least one sentence; and communicating with the client according to the contact mode of the client, and playing the conversation template in the communication process. The problem that the cost is high and the timeliness is poor due to the fact that a customer message needs to be replied by a manual agent when the customer message is idle in the prior art is solved, so that the cost is reduced, the response speed to the customer demand can be improved to a certain extent, and the customer experience is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application. In the drawings:
fig. 1 is a flowchart of an automatic reply method under message semantic analysis according to an embodiment of the present application.
Detailed Description
It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
In this embodiment, an automatic reply method under message semantic analysis is provided, fig. 1 is a flowchart of the automatic reply method under message semantic analysis according to an embodiment of the present application, as shown in fig. 1, and the flowchart includes the following steps:
step S102, obtaining a voice message of a client, wherein the voice message of the client is a voice file which is remained through automatic response after the client performs voice communication;
step S104, converting the voice message into text information;
step S106, classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are pre-configured, and the first category of the text information is at least one of the categories;
there are many ways of classifying, as an alternative to adding, each category is preconfigured with a plurality of corresponding keywords, the text information is matched by using the plurality of keywords, and the number of times of matching the same word in the text is recorded. And recording the occurrence times of the keywords corresponding to the category in the text information for each category, and taking the category with the largest occurrence times as the first category of the text information.
As another alternative to the addition of embodiments, classification may be by way of machine learning. The machine learning model may be trained using multiple sets of training data, each set of training data in the multiple sets of training data including text information and a label corresponding to the text information, the label being used to indicate a category to which the text information belongs. The trained model can be used, text information is input into the model as information, and the label output by the model is the type of the input text information.
As another machine learning classification method, a plurality of sample sentences may be preprocessed to obtain word vector information of each sample sentence, where the word vector information includes a word vector of each word in the sample sentence, and word vectors of partial words in a first sample sentence in the plurality of sample sentences are modified, and each sample sentence corresponds to a target classification result. Classifying each sample sentence based on the sentence classification model and the word vector information of each sample sentence to obtain the prediction classification result of each sample sentence. And acquiring a first loss value based on the prediction classification result and the target classification result. Based on the first loss value, model parameters of the sentence classification model are adjusted. According to the method, parameters of the machine learning module can be adjusted, on one hand, when word vector information corresponding to a sample sentence is obtained, word vectors of partial words in the word vector information are changed, the changed word vector information is used as a basis for classifying the sample sentence, so that the sentence classification model can accurately classify the changed sample sentence, adaptability and resistance of the sentence classification model to the indefinite change of the sentence are enhanced, and robustness of the sentence classification model is improved. On the other hand, the attention point of the sentence classification model can be changed by changing the word vector of a part of words in the sample sentence, and the sentence classification model can accurately classify the word vector of the part of words in the sample sentence due to the change of the word vector of the part of words in the sample sentence, so that the sentence classification model can accurately classify the global features from the global, the local features cannot be excessively attended, and if the local features are excessively attended, the features of the changed part can be possibly acquired, and the accurate classification cannot be completed. Therefore, the situation of excessive fitting can be avoided, and the prediction accuracy of the sentence classification model is improved.
Step S108, determining a speaking template corresponding to the first category according to the first category, wherein the speaking comprises at least one sentence;
as an alternative implementation, the sentences included in the speech templates may be question sentences or statement sentences. A logic process can be further arranged in the conversation template, the logic process is used for identifying answer content after a client requests a question sentence, the sentence corresponding to the condition including the first keyword is played when the answer content includes the first keyword, and the pre-configured sentence corresponding to the condition not including the first keyword is played when the answer content does not include the first keyword.
As another optional implementation manner, during the communication process, the mood of the client may be obtained, the emotion of the client is judged according to the keyword in the voice of the client and the mood, and if the emotion of the client contains an dissatisfied emotion, the client is converted into an artificial seat, and the priority of the client is adjusted to be the highest.
There are many ways to judge the emotion of a customer, for example, to collect a voice sample. Extracting text information in a voice recording sample, preprocessing the text information, inputting a pre-established semantic emotion anger degree detection model, and outputting semantic anger probability evaluation parameters; and acquiring corresponding language anger probability evaluation parameters according to the voice frequency spectrum information in the voice recording sample. And superposing the semantic anger probability evaluation parameter and the mood anger probability evaluation parameter through a Gaussian mixture model to obtain the anger degree comprehensive score of the voice. As a preferred scheme, the obtaining of the semantic anger probability assessment parameter specifically includes: (1) and segmenting the collected text content. (2) Emotion tendencies are classified into three categories of positive, negative and neutral, and the judgment of the mood polarity in the data is initialized by utilizing a dictionary based on the universality mood tendencies in the traditional method. (3) Based on the word segmentation result of (1), sentence vectorization is performed by using the latest models such as BERT or ERNIE2 (provided by *** or bainu). Semantic features are extracted to form a particular dialogue sentence vector. (4) And carrying out word embedding vectorization on the whole sentence by utilizing an ebedding function in the BERT model. That is, a particular word is translated into a vector of N elements. The basic operation of word embedding vectorization is: the specific method is to obtain 12 layers or more of converter tokens by using a Google bi-directional encoder model, then add the vectorized words obtained by the last 3-4 layers, and finally obtain the vectorized representation of the words. For example: "why do you let me wait so long? "such a word can become a matrix after the vectorization change, thereby entering the next machine learning. (5) Training the matrix obtained in the step (4) and the training set obtained in the step (2) by using a deep neural network DNN.
And step S110, communicating with the client according to the contact way of the client, and playing the conversation template in the communication process.
As another optional implementation manner, in the process of playing the speaking template, a record can be further recorded on the answer of the client to the speaking template, then the record is converted into characters, semantic analysis is performed on the characters, whether the client obtains the expected answer is judged, and a label is marked, wherein the label is used for indicating whether the client obtains the expected answer. If the marked label is used for indicating that the client obtains the expected answer, counting the duty ratio of the label which does not obtain the expected answer, and if the duty ratio exceeds 50%, sending the conversation template to an administrator, and prompting that the conversation module needs to be modified. And for the clients with labels which are not expected answers, pushing the records to sales personnel corresponding to the clients, judging whether the labels are correct by the sales personnel, if so, carrying out client follow-up by the sales personnel, and if so, modifying the labels to obtain the expected answers by the sales personnel.
The problems of high cost and poor timeliness caused by the fact that a customer leaves a message and needs a manual agent to reply when the customer leaves a message in the prior art are solved through the steps, so that the cost is reduced, the response speed to the customer demand can be improved to a certain extent, and the customer experience is improved.
In the above step, the first category may include a plurality of categories, and in the case that the first category includes a plurality of categories, a conversation template corresponding to each of the plurality of categories included in the first category may be obtained; and sequencing the obtained conversation templates corresponding to each category to obtain the order of playing the conversation templates by the automatic answering machine.
In the case of including a plurality of categories, the introduction voice of the speaking template corresponding to each category can be obtained first, wherein the introduction voice is used for introducing the content of the speaking template; playing the introduction voice of the conversation template corresponding to each category to the client; receiving a conversation template selected by the customer; the conversation template selected by the client is played first during the communication.
After the conversation template selected by the client is played first in the communication process, if the client requirement is met, the client can end the communication process, so that whether the client ends the communication process can be judged, and under the condition that the client does not end the communication process, the conversation templates other than the conversation template selected by the client are played according to the sequence. Further, after the conversation template is played in the communication process, judging whether the client finishes the communication process, if the client does not finish the communication process, converting the client into an artificial agent, and adjusting the client priority to be the highest.
In this embodiment, there is provided an electronic device including a memory in which a computer program is stored, and a processor configured to run the computer program to perform the method in the above embodiment.
The above-described programs may be run on a processor or may also be stored in memory (or referred to as computer-readable media), including both permanent and non-permanent, removable and non-removable media, and information storage may be implemented by any method or technique. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
These computer programs may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks and/or block diagram block or blocks, and corresponding steps may be implemented in different modules. In this embodiment, there is provided an apparatus called an automatic reply apparatus under message semantic analysis, including: the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring a voice message of a client, wherein the voice message of the client is a voice file which is remained through automatic response after the client performs voice communication; the conversion module is used for converting the voice message into text information; the classification module is used for classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are pre-configured, and the first category of the text information is at least one of the categories; a determining module, configured to determine a speech template corresponding to the first category according to the first category, where the speech includes at least one sentence; and the playing module is used for communicating with the client according to the contact way of the client and playing the conversation template in the communication process.
The modules in the apparatus correspond to the steps in the above method, and have been described in the method, and are not described herein.
For example, the determining module is configured to: obtaining a conversation template corresponding to each of a plurality of categories included in the first category; and sequencing the obtained conversation templates corresponding to each category to obtain the order of playing the conversation templates by the automatic answering machine.
For another example, the playing module is configured to: acquiring introduction voice of a speaking template corresponding to each category, wherein the introduction voice is used for introducing the content of the speaking template; playing the introduction voice of the conversation template corresponding to each category to the client; receiving a conversation template selected by the customer; the conversation template selected by the client is played first during the communication. Optionally, the playing module is further configured to determine whether the client ends the communication process, and play out, according to the order, other voice templates than the voice template selected by the client if the client does not end the communication process. Optionally, the apparatus may further include: and the switching module is used for judging whether the client finishes the communication process after the conversation template is played in the communication process, if not, switching the client into an artificial seat and adjusting the priority of the client to be the highest.
This embodiment can be used in the following scenario: customer information is obtained through a marketing means and actively externally connected to customers, and voice messages of the customers are collected through automatic questions and answers. Then, setting different matched corresponding speaking templates according to the voice message scene, for example, if lottery prizes are not picked up, the matched prizes are picked up speaking templates.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.
Claims (10)
1. An automatic reply method under the semantic analysis of a message is characterized by comprising the following steps:
obtaining a voice message of a client, wherein the voice message of the client is a voice file which is left by automatic response after voice communication is carried out on the client;
converting the voice message into text information;
classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are preconfigured, and the first category of the text information is at least one of the categories; the method comprises the steps of classifying by means of machine learning, wherein a machine learning model is trained by using a plurality of sets of training data, each set of training data in the plurality of sets of training data comprises text information and a label corresponding to the text information, and the label is used for indicating the category to which the text information belongs; the model obtained after training can be used, text information is input into the model as information, and the label output by the model is the type of the input text information; wherein training the machine learning model comprises: preprocessing a plurality of sample sentences to obtain word vector information of each sample sentence, wherein the word vector information comprises word vectors of each word in each sample sentence, the word vectors of partial words in a first sample sentence in the plurality of sample sentences are changed, and each sample sentence corresponds to a target classification result; classifying each sample sentence based on the sentence classification model and the word vector information of each sample sentence to obtain a prediction classification result of each sample sentence; acquiring a first loss value based on the prediction classification result and the target classification result; based on the first loss value, adjusting model parameters of the sentence classification model;
determining a speaking template corresponding to the first category according to the first category, wherein the speaking comprises at least one sentence;
according to the contact mode of the client, communicating with the client, and playing the conversation template in the communication process; recording answers of a customer aiming at the voice template in the process of playing the voice template, converting the recordings into characters, carrying out semantic analysis on the characters, judging whether the customer obtains expected answers or not, and marking tags, wherein the tags are used for indicating whether the customer obtains the expected answers or not; if the marked label is used for indicating the client to obtain the expected answer, counting the duty ratio of the label which does not obtain the expected answer, and if the duty ratio exceeds 50%, sending the speaking template to an administrator, and prompting that the speaking module needs to be modified; and for the clients with labels which are not expected answers, pushing the records to sales personnel corresponding to the clients, judging whether the labels are correct by the sales personnel, if so, carrying out client follow-up by the sales personnel, and if so, modifying the labels to obtain the expected answers by the sales personnel.
2. The method of claim 1, wherein, in the case where the first category includes a plurality of categories, determining from the first category that the first category corresponds to a conversation template includes:
obtaining a conversation template corresponding to each of a plurality of categories included in the first category;
and sequencing the obtained conversation templates corresponding to each category to obtain the order of playing the conversation templates.
3. The method of claim 2, wherein playing the conversation template during the communication comprises:
acquiring introduction voice of a speaking template corresponding to each category, wherein the introduction voice is used for introducing the content of the speaking template;
playing the introduction voice of the conversation template corresponding to each category to the client;
receiving a conversation template selected by the customer;
the conversation template selected by the client is played first during the communication.
4. A method according to claim 3, wherein after first playing the ticket template selected by the customer during the communication, the method further comprises:
judging whether the client finishes the communication process, and playing other voice templates except the voice template selected by the client according to the sequence under the condition that the client does not finish the communication process.
5. The method of any of claims 1-4, wherein after playing the conversation template during the communication, the method further comprises:
judging whether the client finishes the communication process, if not, converting the client into an artificial seat, and adjusting the client priority to be the highest.
6. An automatic reply device under the semantic analysis of a message, comprising:
the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring a voice message of a client, wherein the voice message of the client is a voice file which is remained through automatic response after the client performs voice communication;
the conversion module is used for converting the voice message into text information;
the classification module is used for classifying the text information according to the content of the text information to obtain a first category corresponding to the text information, wherein a plurality of categories are pre-configured, and the first category of the text information is at least one of the categories; the method comprises the steps of classifying by means of machine learning, wherein a machine learning model is trained by using a plurality of sets of training data, each set of training data in the plurality of sets of training data comprises text information and a label corresponding to the text information, and the label is used for indicating the category to which the text information belongs; the model obtained after training can be used, text information is input into the model as information, and the label output by the model is the type of the input text information; wherein training the machine learning model comprises: preprocessing a plurality of sample sentences to obtain word vector information of each sample sentence, wherein the word vector information comprises word vectors of each word in each sample sentence, the word vectors of partial words in a first sample sentence in the plurality of sample sentences are changed, and each sample sentence corresponds to a target classification result; classifying each sample sentence based on the sentence classification model and the word vector information of each sample sentence to obtain a prediction classification result of each sample sentence; acquiring a first loss value based on the prediction classification result and the target classification result; based on the first loss value, adjusting model parameters of the sentence classification model;
a determining module, configured to determine a speech template corresponding to the first category according to the first category, where the speech includes at least one sentence;
the playing module is used for communicating with the client according to the contact way of the client, and playing the conversation template in the communication process; recording answers of a customer aiming at the voice template in the process of playing the voice template, converting the recordings into characters, carrying out semantic analysis on the characters, judging whether the customer obtains expected answers or not, and marking tags, wherein the tags are used for indicating whether the customer obtains the expected answers or not; if the marked label is used for indicating the client to obtain the expected answer, counting the duty ratio of the label which does not obtain the expected answer, and if the duty ratio exceeds 50%, sending the speaking template to an administrator, and prompting that the speaking module needs to be modified; and for the clients with labels which are not expected answers, pushing the records to sales personnel corresponding to the clients, judging whether the labels are correct by the sales personnel, if so, carrying out client follow-up by the sales personnel, and if so, modifying the labels to obtain the expected answers by the sales personnel.
7. The apparatus of claim 6, wherein the means for determining is configured to:
obtaining a conversation template corresponding to each of a plurality of categories included in the first category;
and sequencing the obtained conversation templates corresponding to each category to obtain the order of the conversation templates.
8. The apparatus of claim 7, wherein the play module is configured to:
acquiring introduction voice of a speaking template corresponding to each category, wherein the introduction voice is used for introducing the content of the speaking template;
playing the introduction voice of the conversation template corresponding to each category to the client;
receiving a conversation template selected by the customer;
the conversation template selected by the client is played first during the communication.
9. The apparatus of claim 8, wherein the device comprises a plurality of sensors,
the playing module is further configured to determine whether the client ends the communication process, and play other voice templates except the voice template selected by the client according to the sequence when the client does not end the communication process.
10. The apparatus according to any one of claims 6 to 9, further comprising:
and the switching module is used for judging whether the client finishes the communication process after the conversation template is played in the communication process, if not, switching the client into an artificial seat and adjusting the priority of the client to be the highest.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110831527.9A CN113705186B (en) | 2021-07-22 | 2021-07-22 | Automatic reply method and device under message semantic analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110831527.9A CN113705186B (en) | 2021-07-22 | 2021-07-22 | Automatic reply method and device under message semantic analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113705186A CN113705186A (en) | 2021-11-26 |
CN113705186B true CN113705186B (en) | 2023-12-22 |
Family
ID=78650690
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110831527.9A Active CN113705186B (en) | 2021-07-22 | 2021-07-22 | Automatic reply method and device under message semantic analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113705186B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106326227A (en) * | 2015-06-17 | 2017-01-11 | 中兴通讯股份有限公司 | Method and device for providing online customer service |
CN109190652A (en) * | 2018-07-06 | 2019-01-11 | 中国平安人寿保险股份有限公司 | It attends a banquet sort management method, device, computer equipment and storage medium |
CN109446304A (en) * | 2018-10-10 | 2019-03-08 | 长沙师范学院 | Intelligent customer service session method and system |
CN110149450A (en) * | 2019-05-22 | 2019-08-20 | 欧冶云商股份有限公司 | Intelligent customer service answer method and system |
CN110895940A (en) * | 2019-12-17 | 2020-03-20 | 集奥聚合(北京)人工智能科技有限公司 | Intelligent voice interaction method and device |
CN112463920A (en) * | 2020-11-25 | 2021-03-09 | 联想(北京)有限公司 | Information response method and device |
CN112507094A (en) * | 2020-12-11 | 2021-03-16 | 润联软件***(深圳)有限公司 | Customer service robot dialogue method based on reinforcement learning and related components thereof |
CN112802568A (en) * | 2021-02-03 | 2021-05-14 | 紫东信息科技(苏州)有限公司 | Multi-label stomach disease classification method and device based on medical history text |
CN113032560A (en) * | 2021-03-16 | 2021-06-25 | 北京达佳互联信息技术有限公司 | Sentence classification model training method, sentence processing method and equipment |
CN113064980A (en) * | 2021-03-22 | 2021-07-02 | 苏宁金融科技(南京)有限公司 | Intelligent question and answer method and device, computer equipment and storage medium |
-
2021
- 2021-07-22 CN CN202110831527.9A patent/CN113705186B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106326227A (en) * | 2015-06-17 | 2017-01-11 | 中兴通讯股份有限公司 | Method and device for providing online customer service |
CN109190652A (en) * | 2018-07-06 | 2019-01-11 | 中国平安人寿保险股份有限公司 | It attends a banquet sort management method, device, computer equipment and storage medium |
CN109446304A (en) * | 2018-10-10 | 2019-03-08 | 长沙师范学院 | Intelligent customer service session method and system |
CN110149450A (en) * | 2019-05-22 | 2019-08-20 | 欧冶云商股份有限公司 | Intelligent customer service answer method and system |
CN110895940A (en) * | 2019-12-17 | 2020-03-20 | 集奥聚合(北京)人工智能科技有限公司 | Intelligent voice interaction method and device |
CN112463920A (en) * | 2020-11-25 | 2021-03-09 | 联想(北京)有限公司 | Information response method and device |
CN112507094A (en) * | 2020-12-11 | 2021-03-16 | 润联软件***(深圳)有限公司 | Customer service robot dialogue method based on reinforcement learning and related components thereof |
CN112802568A (en) * | 2021-02-03 | 2021-05-14 | 紫东信息科技(苏州)有限公司 | Multi-label stomach disease classification method and device based on medical history text |
CN113032560A (en) * | 2021-03-16 | 2021-06-25 | 北京达佳互联信息技术有限公司 | Sentence classification model training method, sentence processing method and equipment |
CN113064980A (en) * | 2021-03-22 | 2021-07-02 | 苏宁金融科技(南京)有限公司 | Intelligent question and answer method and device, computer equipment and storage medium |
Non-Patent Citations (1)
Title |
---|
融合神经网络与电力领域知识的智能客服对话***研究;吕诗宁;张毅;胡若云;沈然;江俊军;欧智坚;;浙江电力(第08期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113705186A (en) | 2021-11-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109582793B (en) | Model training method, customer service system, data labeling system and readable storage medium | |
CN104598445B (en) | Automatically request-answering system and method | |
CN108932945A (en) | A kind of processing method and processing device of phonetic order | |
CN112185358A (en) | Intention recognition method, model training method, device, equipment and medium | |
CN111210842A (en) | Voice quality inspection method, device, terminal and computer readable storage medium | |
CN113297365B (en) | User intention judging method, device, equipment and storage medium | |
CN112364622B (en) | Dialogue text analysis method, device, electronic device and storage medium | |
CN114782054A (en) | Customer service quality detection method based on deep learning algorithm and related equipment | |
CN114416989A (en) | Text classification model optimization method and device | |
CN113051384B (en) | User portrait extraction method based on dialogue and related device | |
CN111916110B (en) | Voice quality inspection method and device | |
CN111414732A (en) | Text style conversion method and device, electronic equipment and storage medium | |
CN111739537B (en) | Semantic recognition method and device, storage medium and processor | |
CN113705186B (en) | Automatic reply method and device under message semantic analysis | |
CN116932735A (en) | Text comparison method, device, medium and equipment | |
CN116976341A (en) | Entity identification method, entity identification device, electronic equipment, storage medium and program product | |
CN112687296B (en) | Audio disfluency identification method, device, equipment and readable storage medium | |
CN116150313A (en) | Data expansion processing method and device | |
CN113672715A (en) | Intention recognition system and method | |
CN113705248B (en) | Method and device for processing tactical training data based on result evaluation | |
CN113946673B (en) | Semantic-based intelligent customer service routing processing method and device | |
CN117744661B (en) | Text generation model training method and text generation method based on prompt word engineering | |
CN117453895B (en) | Intelligent customer service response method, device, equipment and readable storage medium | |
CN113723113A (en) | Conversation-based client identity recognition processing method and system | |
CN113255361B (en) | Automatic voice content detection method, device, equipment and 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |