CN111916110A - Voice quality inspection method and device - Google Patents
Voice quality inspection method and device Download PDFInfo
- Publication number
- CN111916110A CN111916110A CN202010785633.3A CN202010785633A CN111916110A CN 111916110 A CN111916110 A CN 111916110A CN 202010785633 A CN202010785633 A CN 202010785633A CN 111916110 A CN111916110 A CN 111916110A
- Authority
- CN
- China
- Prior art keywords
- quality inspection
- target
- voice
- statement
- quality
- 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.)
- Granted
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 332
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000005516 engineering process Methods 0.000 claims abstract description 23
- 238000012163 sequencing technique Methods 0.000 claims abstract description 11
- 238000012372 quality testing Methods 0.000 claims description 23
- 238000003908 quality control method Methods 0.000 claims description 17
- 238000010276 construction Methods 0.000 claims description 11
- 230000000694 effects Effects 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims 5
- 230000008569 process Effects 0.000 description 13
- 238000004590 computer program Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- 238000000605 extraction Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 3
- 206010028980 Neoplasm Diseases 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000036210 malignancy Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000013519 translation Methods 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
- 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
- G10L25/60—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 measuring the quality of voice signals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- 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)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Machine Translation (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a voice quality inspection method and device. The invention comprises the following steps: constructing a quality inspection map according to the data of a target customer, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are associated in the quality inspection map by sequencing in sequence; acquiring target voice to be subjected to quality inspection; converting the target voice into a target text; and inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice. The invention solves the technical problem that the quality inspection efficiency is lower because the quality inspection of the associated quality inspection points in the recording can not be carried out when the related technology carries out the quality inspection on the telephone recording.
Description
Technical Field
The invention relates to the field of voice recognition, in particular to a voice quality inspection method and device.
Background
The quality inspection of telephone is to inspect the quality of telephone recordings for sale and check the illegal points. In the related art, the conventional telephone quality inspection generally performs quality inspection on a text after voice translation by using a regular matching quality inspection technology. Generally, the problem of single isolated quality inspection is inspected, joint quality inspection cannot be carried out on related quality inspection points, and the problem of low quality inspection accuracy of certain quality inspection items exists. If the customer is asked to ask that the purchase intention is a quality inspection point, and the verification user information, such as the ID number, is also a quality inspection point, but the quality inspection of the customer information and the verification user information is related, namely if the first quality inspection point is illegal, and the second quality inspection point is in compliance under an isolated judgment, the final result is illegal. Therefore, the accuracy of the telephone quality inspection in the related art is low.
In view of the above problems in the related art, no effective solution has been proposed.
Disclosure of Invention
The invention mainly aims to provide a voice quality inspection method and a voice quality inspection device, which are used for solving the technical problem of low quality inspection efficiency caused by the fact that related quality inspection points in a recording cannot be subjected to quality inspection when the related technology is used for performing quality inspection on a telephone recording.
To achieve the above object, according to one aspect of the present invention, a method for voice quality inspection is provided. The invention comprises the following steps: constructing a quality inspection map according to the data of a target customer, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are associated in the quality inspection map by sequencing in sequence; acquiring target voice to be subjected to quality inspection; converting the target voice into a target text; and inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice.
Further, after acquiring the target voice to be quality-checked, the method further comprises: judging the number of source sound channels of the target voice; if the source sound channel is one, the target voice is segmented by adopting a voice activity detection technology and a voiceprint recognition technology so as to segment the voice corresponding to the target client and the voice corresponding to the target seat; respectively storing the voice corresponding to the target customer and the voice corresponding to the target seat in different sound channels; and if the number of the source channels is two, determining the voice corresponding to the target customer and the voice corresponding to the target seat through the identity tag corresponding to the source channel.
Further, before inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice, the method comprises the following steps: dividing the target text into a plurality of clauses; acquiring keywords required by quality inspection of target voice; and converting the plurality of clauses into a preset inquiry statement and a preset reply statement through the keywords and the plurality of clauses.
Further, before inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice, the method comprises the following steps: dividing the target text into a plurality of clauses; and filtering the attributes of the quality inspection entities contained in the plurality of clauses through regular matching so as to convert the plurality of clauses into a preset inquiry statement and a preset reply statement.
Further, the plurality of quality inspection entities at least comprise a first quality inspection entity and a second quality inspection entity, the first quality inspection entity is sequenced before the second quality inspection entity, and the step of inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice comprises the following steps: inserting a preset inquiry statement and a preset reply statement into a quality inspection map; if the preset inquiry statement and the preset reply statement both have the statement corresponding to the first quality inspection entity, performing quality inspection on the content included in the second quality inspection entity according to the preset inquiry statement and the preset reply statement; if the preset inquiry statement has a statement corresponding to the first quality control entity, and the preset reply statement has no statement corresponding to the first quality control entity, ending the quality control of the target voice; and if the preset inquiry statement and the preset reply statement do not have the statement corresponding to the first quality inspection entity, ending the quality inspection of the target voice.
In order to achieve the above object, according to another aspect of the present invention, there is provided an apparatus for voice quality inspection. The device includes: the construction unit is used for constructing a quality inspection map according to the data of the target client, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are related in the quality inspection map by sequencing in sequence; the first acquisition unit is used for acquiring target voice to be subjected to quality inspection; the first conversion unit is used for converting the target voice into a target text; and the quality inspection unit is used for inserting the content in the target text into a quality inspection map so as to perform quality inspection on the target voice.
To achieve the above object, according to another aspect of the present invention, there is provided a computer-readable storage medium. The computer readable storage medium includes a stored program, wherein the program controls an apparatus in which the computer readable storage medium is located to execute the above-mentioned voice quality inspection method when the program runs.
To achieve the above object, according to another aspect of the present invention, there is provided a processor. The processor is used for running a program, wherein the program executes the voice quality inspection method.
The invention adopts the following steps: constructing a quality inspection map according to the data of a target customer, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are associated in the quality inspection map by sequencing in sequence; acquiring target voice to be subjected to quality inspection; converting the target voice into a target text; the content in the target text is inserted into the quality inspection map to perform quality inspection on the target voice, so that the technical problem that the quality inspection efficiency is low due to the fact that the quality inspection cannot be performed on the associated quality inspection points in the recording when the quality inspection is performed on the telephone recording in the related technology is solved, and the technical effect of improving the telephone quality inspection efficiency is achieved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart of a voice quality inspection method according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating the voice quality inspection method according to the embodiment of the present invention after step S102 in fig. 1;
fig. 3 is a flowchart illustrating a voice quality inspection method of the voice quality inspection method according to an embodiment of the present invention before step S104 in fig. 1;
fig. 4 is another flow chart illustrating a voice quality inspection method of the voice quality inspection method according to an embodiment of the present invention before step S104 in fig. 1;
fig. 5 is a flowchart of the voice quality inspection method according to the embodiment of the present invention, specifically illustrating step S104 in fig. 1;
fig. 6 is a schematic diagram of an apparatus for voice quality inspection according to an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection 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 may be interchanged under appropriate circumstances in order to facilitate the description of the embodiments of the invention 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, system, article, or apparatus 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 apparatus.
Example 1
In accordance with an embodiment of the present invention, there is provided a method embodiment for voice quality inspection, it is noted that the steps illustrated in the flowchart of the drawings 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 flowchart, in some cases the steps illustrated or described may be performed in an order different than that described herein.
Fig. 1 is a flowchart illustrating a voice quality inspection method according to an embodiment of the present invention. As shown in fig. 1, the present invention comprises the steps of:
step S101, constructing a quality inspection map according to the data of a target client, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are related in the quality inspection map by sequencing;
the application provides a method for carrying out quality inspection on telemarketing sound records by constructing a knowledge graph, wherein the knowledge graph is specifically required to be constructed according to the data of a target client.
It should be noted that the knowledge graph provided by the embodiment of the present application is used for describing concepts, entities, events and relationships among the concepts, and serves as a core foundation for building a next-generation intelligent search engine. Generally, a knowledge graph is a relational network obtained by connecting all different kinds of information together.
In other words, a knowledge graph is a structured semantic knowledge base that describes concepts and their interrelationships in the physical world in symbolic form. The basic composition unit is an entity-relation-entity triple, entities and related attribute-value pairs thereof, and the entities are mutually connected through relations to form a network knowledge structure. The conversion from the webpage link to the concept link of the Web can be realized through the knowledge map, and the retrieval of a user according to a theme instead of a character string is supported, so that the semantic retrieval is really realized. The knowledge graph-based search engine can feed back structured knowledge to a user in a graph mode, and the user can accurately position and deeply acquire the knowledge without browsing a large number of webpages.
The construction of the knowledge graph, also called knowledge modeling, refers to what way to express knowledge, and the core of the construction is to construct an ontology to describe target knowledge.
The ontology is a knowledge description framework, and defines a knowledge category system, concepts and entities under each category, attributes of a certain category of concepts and entities, and semantic relationships between concepts and entities.
There are two methods for the construction of the knowledge system:
1. one is top-down: firstly, constructing a perfect knowledge system, and then filling knowledge into the knowledge system;
2. the other is from bottom to top: namely, in the knowledge extraction process, the knowledge system is automatically expanded and constructed. The bottom-up approach is currently more popular.
Currently, data in the knowledge graph is described using an RDF data model, that is, a triple schema, i.e., "entity 1" - "relationship" - "entity 2". For example, the triplets, "chairman (mazechu, chinese peace)", "chairman" indicates the relationship, "mazechu" is the head entity, and "chinese peace" is the tail entity.
At present, most of knowledge graphs are constructed in a bottom-up mode, and the general construction process of the knowledge graphs is a process of continuous iteration updating.
Specifically, in an alternative embodiment of the present application, a method for quality control of conversational speech between a client and an agent is provided, so that a knowledge graph for processing speech to be quality-controlled is first constructed from the target client's knowledge.
Generally, quality inspection of a speech to be inspected includes at least the following entities: the purchase intention (unknown) of the customer, the name (known) of the customer, the mobile phone number (known) of the customer and the identity card number (known) of the customer are associated with the quality inspection entities by constructing a quality inspection map, and the quality inspection sequence of different entities is divided in the process of performing quality inspection on the voice to be inspected.
In the application scenario provided in this embodiment, the entities that need to be constructed in the knowledge graph include:
1. customer purchase intent entity: the client telephone reads the purchase intention attribute;
2. order number entity: an order number attribute;
3. customer name entity: order number attribute, real name attribute of client, name attribute checked by client service, name attribute read in client telephone, and name check result attribute;
4. customer mobile phone number entity: the method comprises the following steps that (1) the attribute of a booking number, the attribute of a real mobile phone number of a client, the attribute of the front 7 digits of the mobile phone number checked by a customer service, the attribute of the back 4 digits of the mobile phone number read by the telephone of the client and the attribute of the checking result of the mobile phone number are checked;
5. customer identification card number entity: the attribute of the order number, the attribute of the real identity card number of the client, the attribute of the first 14 digits of the identity card number checked by the customer service, the attribute of the last 4 digits of the identity card number read by the telephone of the client and the attribute of the result checked by the identity card number.
In the above, the relevance between the quality inspection entities is embodied by constructing the quality knowledge graph, wherein the quality inspection relevance comprises: 1) four checks (name, identification card, phone number, purchase intention) must be done for each order number; 2) if the purchase intention is negative, the customer service can not carry out four checks (name, identity card, mobile phone number and purchase intention).
The correlation relationship of the manual conversion into the knowledge graph base is as follows:
1) confirming the customer purchase intention entity by order number entity (first check); 2) checking the customer name entity by the customer purchase intention entity (second check);
3) checking the customer mobile phone number entity through a customer purchase intention entity (second check); 4) the customer identification number entity is checked by the customer purchase intention entity (second check).
Therefore, by constructing the knowledge graph of the quality testing entity with the quality testing sequence, the quality testing process is shortened, and the quality testing efficiency is improved.
Step S102, obtaining target voice to be quality checked;
specifically, a target voice to be quality-checked is obtained, and in general, the voice to be quality-checked includes two roles, namely a client and an agent (also called a client specialist).
It should be noted that, if there are a plurality of roles included in the voice to be quality-checked, a scheme of dividing different roles by dividing the roles in the voice to be quality-checked also belongs to the limited scope of the present application, and therefore, a scheme of including a plurality of roles in the voice to be quality-checked is not described in detail here, and the target voice including two roles is exemplified in the present embodiment.
In an alternative embodiment, referring to fig. 2, which is a schematic flow chart after step S102 in fig. 1 of the voice quality inspection method provided in the embodiment of the present invention, after the target voice to be inspected is acquired, the method further includes:
in step S201, the number of source channels of the target voice is determined.
As described above, since the target speech includes at least two characters, the step of determining the character is performed after the target speech is acquired.
First, the number of source channels of the target voice needs to be determined.
Specifically, in general, when only two roles, i.e., a client and an agent, are included in the target speech, the source channels of the target speech are divided into one or two channels.
Step S202, if the source sound channel is one, the target voice is segmented by adopting a voice activity detection technology and a voiceprint recognition technology so as to segment the voice corresponding to the target client and the voice corresponding to the target seat.
Step S203, storing the voice corresponding to the target customer and the voice corresponding to the target agent in different channels.
In one case, if the target voice is recorded in a mono channel, the voice is segmented using VAD and voiceprint recognition techniques, and the segmented voices are stored in different channels, respectively.
Step S204, if the number of the source sound channels is two, the voice corresponding to the target customer and the voice corresponding to the target seat are determined through the identity tag corresponding to the source sound channel.
In another case, the telephone quality inspection is based on the telephone scene, and generally adopts the dual-channel recording, the customer voice is recorded to the sound channel 1, the seat voice is recorded to the sound channel 2, the voices of the customer and the customer service are not interfered with each other, and the telephone quality inspection has a role identity label, so that the user can clearly and definitely know that the source of the voice is the customer or the seat.
Through the steps, the voices corresponding to different roles contained in the target voice are clearly distinguished before the target voice quality inspection, and the problem of poor quality inspection effect caused by unclear voice roles is solved when the target voice is subsequently subjected to quality inspection.
Step S103, converting the target voice into a target text;
specifically, after voice recognition is performed on the voices of the client and the seat, the target voice is converted into a target text to obtain text content corresponding to the voices.
It should be noted that the speech recognition used when converting the target speech into the target text may use an open source framework, or may use an existing API, and the specific means thereof is not limited to other specific means.
And step S104, inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice.
In an alternative embodiment, referring to fig. 3, a flowchart of a method for voice quality inspection according to an embodiment of the present invention before step S104 in fig. 1 is shown, before inserting contents in a target text into a quality inspection map to perform quality inspection on a target voice, the method includes the following steps:
in step S301, the target text is divided into a plurality of clauses.
In the above, after the target voice to be quality-checked is converted into the target text, an intention recognition stage is entered, and the first step is to convert the target text corresponding to the target customer and the target seat into a plurality of clauses with relatively short lengths.
Step S302, obtaining keywords corresponding to the quality inspection of the target voice.
Specifically, the intention of the text content obtained by speech recognition of the speech corresponding to the customer and the agent is judged. Here it is intended to judge the classification of question sentences into different categories. As a common problem in the examples: "do you suffer from malignancy? "do your name? Xxxx "", the first 14 digits of your identification number are xxxx, the last four digits are? "check by the above-mentioned phrase.
The method for judging the intention is a keyword matching technology, and the method comprises the steps of manually classifying question sentences of various types, and extracting keywords of the question sentence types respectively. The question keywords for asking health classes are typically [ 'suffering from', 'tumour' ]. The general keywords when interrogating a document are [ 'identity card', '14 digit' ]. When performing word matching, if a question contains a keyword of a question of a certain category, it is determined as belonging to the question of the certain category.
Step S303, converting the clauses into a preset query statement and a preset reply statement by using the keyword and the clauses.
Because different clients and different agents have different specific utterances corresponding to the query and the answer, the clauses are converted into the preset query sentence and the preset reply sentence which are associated with the quality control knowledge graph through the matching of the keywords,
for example, agent 1 queries for name are: your name is XX, and for do, the inquiry of seat 2 for name is: take your minutes, need check your name, ask you be XX. Therefore, although the specific query sentences of the agent 1 and the agent 2 are different, the present embodiment can recognize the intention of the agent as "checking the name of the customer" by the keyword "name" two words, recognize the intention as "checking the name intention" when one of the above-mentioned persons is encountered by the keyword recognition, and convert the intention into the preset query sentence for quality inspection by the recognized intention, for example: "your name is … …".
It should be noted that, by identifying the intention of the customer or the agent, the identified intention is converted into a preset query statement and a preset reply statement corresponding to the quality inspection in the knowledge graph, where the preset query statement and the preset reply statement may be specifically set according to actual situations in different scenarios.
In summary, no matter how different specific inquiry sentences of an agent are, the embodiment can identify the intention of the agent or the customer by matching keywords, and convert clauses including corresponding inquiry keywords into preset inquiry sentences or preset reply sentences.
In another alternative embodiment, referring to fig. 4, another schematic flow chart before step S104 in fig. 1 of the method for voice quality inspection according to the method for voice quality inspection provided by the embodiment of the present invention is shown, before inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice, the method includes the following steps:
in step S401, the target text is divided into a plurality of clauses.
Step S402, filtering the attributes of the quality inspection entities contained in the clauses through regular matching so as to convert the clauses into a preset inquiry statement and a preset reply statement.
In another technical means for intention recognition provided by the present application, the attributes of the quality inspection entities included in the multiple sub-tools are filtered to obtain corresponding preset query statements and preset reply statements.
In particular, Named Entity Recognition (NER), also referred to as entity recognition, entity segmentation and entity extraction, is a subtask of information extraction that aims to locate and classify named entities in text into predefined categories such as people, organizations, locations, temporal expressions, quantities, monetary values and percentages, etc.
For example, in the present embodiment, the name of a person, the order number, the identification number, the mobile phone number, and the like may be identified and extracted by named entity identification. In addition, the specific attributes of the entities can be filtered through regular matching, and the purpose of extraction can be achieved as well, such as identification number, total 14 digits, and extraction can be carried out by using "\ d {14 }".
Through the extraction, the preset inquiry sentences and the preset reply sentences corresponding to the clauses are identified, and the purpose of identifying the intentions of the seats or the clients can be achieved through the converted preset inquiry sentences and the converted preset reply sentences.
It should be noted that, through the above step of recognizing the intention, the intention in the text corresponding to the customer or the agent may also be directly recognized and marked, and there is no need to convert the recognized intention into a preset query sentence or a preset reply sentence, for example, when recognizing through a keyword, the intention of recognizing the agent is: when the customer is asked for the identification number, the intention of "asking for the identification number" is displayed by a label.
In an alternative embodiment, referring to fig. 5, a detailed flowchart of step S104 in fig. 1 of a voice quality inspection method of the voice quality inspection method provided in the embodiment of the present invention is shown, where the plurality of quality inspection entities at least include a first quality inspection entity and a second quality inspection entity, the first quality inspection entity is ordered before the second quality inspection entity, and inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice includes:
step S501, inserting a preset inquiry statement and a preset reply statement into a quality inspection map.
Firstly, after presetting a preset inquiry statement and a preset reply statement, the preset inquiry statement and the preset reply statement are associated and mapped with the constructed knowledge map.
In the above, the information to be inserted into the knowledge graph may be obtained according to the above steps, where the information includes a preset query statement or a preset reply statement reflecting the intention of the character, and the essence of the information is conversion from a human natural language to a database language.
Step S502, if the preset inquiry sentence and the preset reply sentence have the sentence corresponding to the first quality inspection entity, performing quality inspection on the content included in the second quality inspection entity according to the preset inquiry sentence and the preset reply sentence.
Specifically, after a preset inquiry statement and a preset reply statement are inserted into the constructed knowledge graph, a quality inspection entity with a sequence order in the knowledge graph library is used for carrying out inquiry quality inspection on the preset inquiry statement and the preset reply statement.
In this embodiment, the first quality inspection entity is the purchase intention of the customer, and the second quality inspection entity at least includes the following three items: name, identification card, mobile phone number.
Specifically, if the preset inquiry statement and the preset reply statement inserted into the knowledge graph both have statements corresponding to the purchase intention of the customer, it is indicated that the customer has the purchase intention for the agent inquiry, therefore, when the customer has the purchase intention, Ture is output for the purchase intention of the customer, and quality inspection is continuously performed on entities in the second quality inspection entity, wherein the quality inspection sequence of the entities contained in the second quality inspection entity is not in sequence.
For example: the preset inquiry statement corresponding to the purchase intention of the customer is 'whether the customer intends to buy XX', the preset reply statement corresponding to the purchase intention of the customer is 'intention', if the statements corresponding to the purchase intention exist in the preset inquiry statement and the preset reply statement, the customer is proved to have the purchase intention for the product, and the quality inspection of the entity in the second quality inspection entity can be continuously carried out.
Step S503, if there is a sentence corresponding to the first quality inspection entity in the preset query sentence and there is no sentence corresponding to the first quality inspection entity in the preset reply sentence, ending the quality inspection of the target voice.
In another case, if it is known that the agent inquires about the purchase intention of the customer correspondingly through the preset inquiry statement, but the customer does not reply correspondingly, it is indicated that the customer does not have the purchase intention, it is also determined that the quality inspection of the target voice does not pass, and the quality inspection process of the target voice is finished, and it is not necessary to perform quality inspection on the quality inspection entity in the second quality inspection entity, that is, it is not necessary to perform quality inspection on any item of the name, the identity card, and the mobile phone number.
Alternatively, when it is determined that the customer has no purchase intention through the preset reply statement, quality inspection may be performed on the quality inspection entities included in the second quality inspection entities, but even if any one of the second quality inspection entities outputs a result attribute of Ture, the quality inspection result is not passed.
Step S504, if the preset inquiry statement and the preset reply statement do not have the statement corresponding to the first quality inspection entity, the quality inspection of the target voice is finished.
Further, after quality inspection is carried out on the quality inspection entity through the knowledge graph, the truer False result is directly judged on the result of the association relation. Such as: and performing quality inspection on the purchase intention according to the sequence of the quality inspection entities, wherein the quality inspection entities in the knowledge graph have corresponding relations with the preset inquiry sentences and the preset reply sentences, and if the preset inquiry sentences and the preset reply sentences do not have sentences corresponding to the purchase intention of the customer, judging that the quality inspection of the target voice does not pass, and immediately finishing the quality inspection process of the target voice.
For example: if the preset inquiry statement corresponding to the purchase intention of the customer is 'XX' or 'XX', the preset reply statement corresponding to the purchase intention of the customer is 'intentional', and if the preset inquiry statement and the preset reply statement do not have the statement corresponding to the purchase intention, the situation that the customer directly hangs up the telephone or the customer does not reply to the inquiry of the seat occurs is shown, the quality inspection process can be finished without judging that the target voice quality inspection does not pass, and the quality inspection of the content in the second quality inspection entity is not needed.
Through the quality inspection steps and the construction of the knowledge graph for voice quality inspection, the problem that the quality inspection of entities with the associated relation is difficult in the traditional quality inspection technology is solved. Meanwhile, quality inspection is carried out through the knowledge graph, only different entities, entity attributes and incidence relations need to be created, the information extraction layer and the quality inspection layer are stripped to carry out quality inspection analysis, and the technical problems of low quality inspection efficiency and poor quality inspection effect caused by quality inspection aiming at isolated problems in voice are avoided.
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 different than presented herein. The embodiment of the present invention further provides a device for voice quality inspection, and it should be noted that the device for voice quality inspection according to the embodiment of the present invention can be used for executing the method for voice quality inspection according to the embodiment of the present invention. The following describes a voice quality inspection apparatus according to an embodiment of the present invention.
Example 2
Fig. 6 is a schematic diagram of an apparatus for voice quality inspection according to an embodiment of the present invention. As shown in fig. 6, the apparatus includes: the construction unit 601 is configured to construct a quality inspection map according to the data of the target customer, where the quality inspection map includes a plurality of quality inspection entities, and the quality inspection entities are associated in the quality inspection map by sequencing; a first obtaining unit 602, configured to obtain a target voice to be quality tested; a first conversion unit 603 for converting the target speech into a target text; and the quality inspection unit 604 is used for inserting the content in the target text into a quality inspection map so as to perform quality inspection on the target voice.
The voice quality inspection device provided by the embodiment of the invention is used for constructing a quality inspection map according to the data of a target client through the construction unit 601, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are associated by sequencing in sequence in the quality inspection map; a first obtaining unit 602, configured to obtain a target voice to be quality tested; a first conversion unit 603 for converting the target speech into a target text; the quality inspection unit 604 is configured to insert the content in the target text into a quality inspection map to perform quality inspection on the target voice, so that the technical problem that quality inspection efficiency is low due to the fact that quality inspection cannot be performed on associated quality inspection points in a recording when a telephone recording is subjected to quality inspection in the related art is solved, and the technical effect of improving the telephone quality inspection efficiency is further achieved.
Optionally, the apparatus further comprises: the device comprises a judging unit, a quality control unit and a quality control unit, wherein the judging unit is used for judging the number of source sound channels of target voice after the target voice to be subjected to quality control is acquired; the first segmentation unit is used for segmenting the target voice by adopting a voice activity detection technology and a voiceprint recognition technology under the condition that the source sound channel is one so as to segment the voice corresponding to the target client and the voice corresponding to the target seat; the storage unit is used for respectively storing the voice corresponding to the target customer and the voice corresponding to the target seat in different sound channels; and the first determining unit is used for determining the voice corresponding to the target customer and the voice corresponding to the target seat through the identity tag corresponding to the source channel under the condition that the number of the source channels is two.
Optionally, the apparatus comprises: the second segmentation unit is used for segmenting the target text into a plurality of clauses before inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice; the second acquisition unit is used for acquiring keywords required by quality inspection of the target voice; and the second conversion unit is used for converting the clauses into a preset inquiry statement and a preset reply statement through the keywords and the clauses.
Optionally, the apparatus comprises: the third segmentation unit is used for segmenting the target text into a plurality of clauses before inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice; and the third conversion unit is used for filtering the attributes of the quality inspection entities contained in the multiple clauses through regular matching so as to convert the multiple clauses into a preset inquiry statement and a preset reply statement.
Optionally, the quality inspection entities at least include a first quality inspection entity and a second quality inspection entity, the first quality inspection entity is ordered before the second quality inspection entity, and the quality inspection unit 604 includes: the inserting subunit is used for inserting the preset inquiry sentences and the preset reply sentences into the quality inspection map; the first quality inspection subunit is used for performing quality inspection on the content included in the second quality inspection entity according to the preset inquiry statement and the preset reply statement under the condition that the statements corresponding to the first quality inspection entity exist in the preset inquiry statement and the preset reply statement; the second quality testing subunit is used for finishing the quality testing of the target voice under the condition that the preset inquiry statement has the statement corresponding to the first quality testing entity and the preset reply statement has no statement corresponding to the first quality testing entity; and the third quality testing subunit is used for finishing the quality testing of the target voice under the condition that the preset inquiry statement and the preset reply statement do not have the statement corresponding to the first quality testing entity.
The device for voice quality inspection comprises a processor and a memory, wherein the construction unit 601 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, and the technical problem that the quality inspection efficiency is low because the quality inspection of the associated quality inspection points in the recording cannot be performed when the quality inspection is performed on the telephone recording in the related technology is solved by adjusting the kernel parameters.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
Example 3
An embodiment of the present invention further provides a computer-readable storage medium, on which a program is stored, where the program, when executed by a processor, implements a method for voice quality inspection.
The embodiment of the invention provides a processor, which is used for running a program, wherein a voice quality inspection method is executed when the program runs.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps: constructing a quality inspection map according to the data of a target customer, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are associated in the quality inspection map by sequencing in sequence; acquiring target voice to be subjected to quality inspection; converting the target voice into a target text; and inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice.
Optionally, after acquiring the target voice to be quality-checked, the method further includes: judging the number of source sound channels of the target voice; if the source sound channel is one, the target voice is segmented by adopting a voice activity detection technology and a voiceprint recognition technology so as to segment the voice corresponding to the target client and the voice corresponding to the target seat; respectively storing the voice corresponding to the target customer and the voice corresponding to the target seat in different sound channels; and if the number of the source channels is two, determining the voice corresponding to the target customer and the voice corresponding to the target seat through the identity tag corresponding to the source channel.
Optionally, before inserting the content in the target text into the quality testing map to perform quality testing on the target voice, the method includes: dividing the target text into a plurality of clauses; acquiring keywords required by quality inspection of target voice; and converting the plurality of clauses into a preset inquiry statement and a preset reply statement through the keywords and the plurality of clauses.
Optionally, before inserting the content in the target text into the quality testing map to perform quality testing on the target voice, the method includes: dividing the target text into a plurality of clauses; and filtering the attributes of the quality inspection entities contained in the plurality of clauses through regular matching so as to convert the plurality of clauses into a preset inquiry statement and a preset reply statement.
Optionally, the plurality of quality inspection entities at least include a first quality inspection entity and a second quality inspection entity, the first quality inspection entity is ordered before the second quality inspection entity, and the inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice includes: inserting a preset inquiry statement and a preset reply statement into a quality inspection map; if the preset inquiry statement and the preset reply statement both have the statement corresponding to the first quality inspection entity, performing quality inspection on the content included in the second quality inspection entity according to the preset inquiry statement and the preset reply statement; if the preset inquiry statement has a statement corresponding to the first quality control entity, and the preset reply statement has no statement corresponding to the first quality control entity, ending the quality control of the target voice; and if the preset inquiry statement and the preset reply statement do not have the statement corresponding to the first quality inspection entity, ending the quality inspection of the target voice.
The device herein may be a server, a PC, a PAD, a mobile phone, etc.
Example 4
The invention also provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: constructing a quality inspection map according to the data of a target customer, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are associated in the quality inspection map by sequencing in sequence; acquiring target voice to be subjected to quality inspection; converting the target voice into a target text; and inserting the content in the target text into a quality inspection map to perform quality inspection on the target voice.
Optionally, after acquiring the target voice to be quality-checked, the method further includes: judging the number of source sound channels of the target voice; if the source sound channel is one, the target voice is segmented by adopting a voice activity detection technology and a voiceprint recognition technology so as to segment the voice corresponding to the target client and the voice corresponding to the target seat; respectively storing the voice corresponding to the target customer and the voice corresponding to the target seat in different sound channels; and if the number of the source channels is two, determining the voice corresponding to the target customer and the voice corresponding to the target seat through the identity tag corresponding to the source channel.
Optionally, before inserting the content in the target text into the quality testing map to perform quality testing on the target voice, the method includes: dividing the target text into a plurality of clauses; acquiring keywords required by quality inspection of target voice; and converting the plurality of clauses into a preset inquiry statement and a preset reply statement through the keywords and the plurality of clauses.
Optionally, before inserting the content in the target text into the quality testing map to perform quality testing on the target voice, the method includes: dividing the target text into a plurality of clauses; and filtering the attributes of the quality inspection entities contained in the plurality of clauses through regular matching so as to convert the plurality of clauses into a preset inquiry statement and a preset reply statement.
Optionally, the plurality of quality inspection entities at least include a first quality inspection entity and a second quality inspection entity, the first quality inspection entity is ordered before the second quality inspection entity, and the inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice includes: inserting a preset inquiry statement and a preset reply statement into a quality inspection map; if the preset inquiry statement and the preset reply statement both have the statement corresponding to the first quality inspection entity, performing quality inspection on the content included in the second quality inspection entity according to the preset inquiry statement and the preset reply statement; if the preset inquiry statement has a statement corresponding to the first quality control entity, and the preset reply statement has no statement corresponding to the first quality control entity, ending the quality control of the target voice; and if the preset inquiry statement and the preset reply statement do not have the statement corresponding to the first quality inspection entity, ending the quality inspection of the target voice.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions 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 flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media 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 that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present invention, and are not intended to limit the present invention. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A method for voice quality inspection, comprising:
constructing a quality inspection map according to the data of a target customer, wherein the quality inspection map comprises a plurality of quality inspection entities, and the quality inspection entities are related in the quality inspection map by sequencing in sequence;
acquiring target voice to be subjected to quality inspection;
converting the target voice into a target text;
and inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice.
2. The method of claim 1, wherein after acquiring the target speech to be quality tested, the method further comprises:
judging the number of source sound channels of the target voice;
if the source sound channel is one, the target voice is segmented by adopting a voice activity detection technology and a voiceprint recognition technology so as to segment the voice corresponding to the target client and the voice corresponding to the target seat;
respectively storing the voice corresponding to the target customer and the voice corresponding to the target seat in different sound channels;
and if the number of the source sound channels is two, determining the voice corresponding to the target customer and the voice corresponding to the target seat through the identity tag corresponding to the source sound channel.
3. The method of claim 1, wherein prior to inserting the content in the target text into the quality testing graph for quality testing of the target speech, the method comprises:
dividing the target text into a plurality of clauses;
acquiring keywords required by quality inspection of the target voice;
and converting the clauses into a preset inquiry statement and a preset reply statement through the keyword and the clauses.
4. The method of claim 1, wherein prior to inserting the content in the target text into the quality testing graph for quality testing of the target speech, the method comprises:
dividing the target text into a plurality of clauses;
and filtering the attributes of the quality inspection entities contained in the clauses through regular matching so as to convert the clauses into preset inquiry sentences and preset reply sentences.
5. The method of claim 3 or 4, wherein the plurality of quality inspection entities comprises at least a first quality inspection entity and a second quality inspection entity, the first quality inspection entity is ordered before the second quality inspection entity, and the inserting the content of the target text into the quality inspection map for quality inspection of the target speech comprises:
inserting the preset inquiry statement and the preset reply statement into the quality inspection map;
if the preset inquiry statement and the preset reply statement both have the statement corresponding to the first quality inspection entity, performing quality inspection on the content included in the second quality inspection entity according to the preset inquiry statement and the preset reply statement;
if the preset inquiry statement contains a statement corresponding to the first quality control entity, and the preset reply statement does not contain the statement corresponding to the first quality control entity, ending the quality control of the target voice;
and if the preset inquiry statement and the preset reply statement do not have the statement corresponding to the first quality inspection entity, ending the quality inspection of the target voice.
6. An apparatus for voice quality inspection, comprising:
the system comprises a construction unit, a quality inspection mapping unit and a quality inspection mapping unit, wherein the construction unit is used for constructing a quality inspection mapping according to the data of a target client, the quality inspection mapping comprises a plurality of quality inspection entities, and the quality inspection entities in the quality inspection mapping are associated by sequencing in sequence;
the first acquisition unit is used for acquiring target voice to be subjected to quality inspection;
a first conversion unit for converting the target voice into a target text;
and the quality inspection unit is used for inserting the content in the target text into the quality inspection map so as to perform quality inspection on the target voice.
7. The apparatus of claim 6, further comprising:
the device comprises a judging unit, a processing unit and a processing unit, wherein the judging unit is used for judging the number of source sound channels of target voice after the target voice to be subjected to quality inspection is acquired and before the target voice is converted into a target text;
the first segmentation unit is used for segmenting the target voice by adopting a voice activity detection technology and a voiceprint recognition technology under the condition that the source sound channel is one so as to segment the voice corresponding to the target client and the voice corresponding to the target seat;
the storage unit is used for respectively storing the voice corresponding to the target customer and the voice corresponding to the target seat in different sound channels;
a first determining unit, configured to determine, when the number of source channels is two, a voice corresponding to the target customer and a voice corresponding to the target agent through an identity tag corresponding to the source channel.
8. The apparatus of claim 6, wherein the apparatus comprises:
the second segmentation unit is used for segmenting the target text into a plurality of clauses before inserting the content in the target text into the quality inspection map to perform quality inspection on the target voice;
a second obtaining unit, configured to obtain a keyword required for quality inspection of the target voice;
and the second conversion unit is used for converting the clauses into a preset inquiry statement and a preset reply statement through the keyword and the clauses.
9. A computer-readable storage medium, comprising a stored program, wherein the program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the method of voice quality inspection according to any one of claims 1 to 5.
10. A processor, characterized in that the processor is configured to run a program, wherein the program is configured to perform the method of voice quality inspection according to any one of claims 1 to 5 when running.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010785633.3A CN111916110B (en) | 2020-08-06 | 2020-08-06 | Voice quality inspection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010785633.3A CN111916110B (en) | 2020-08-06 | 2020-08-06 | Voice quality inspection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111916110A true CN111916110A (en) | 2020-11-10 |
| CN111916110B CN111916110B (en) | 2024-04-26 |
Family
ID=73287330
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010785633.3A Active CN111916110B (en) | 2020-08-06 | 2020-08-06 | Voice quality inspection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111916110B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113128231A (en) * | 2021-04-25 | 2021-07-16 | 深圳市慧择时代科技有限公司 | Data quality inspection method and device, storage medium and electronic equipment |
| CN113345430A (en) * | 2021-06-25 | 2021-09-03 | 上海适享文化传播有限公司 | Multi-field query method based on voice fixed condition |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017041372A1 (en) * | 2015-09-07 | 2017-03-16 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| WO2017076263A1 (en) * | 2015-11-03 | 2017-05-11 | 中兴通讯股份有限公司 | Method and device for integrating knowledge bases, knowledge base management system and storage medium |
| US20180060733A1 (en) * | 2016-08-31 | 2018-03-01 | International Business Machines Corporation | Techniques for assigning confidence scores to relationship entries in a knowledge graph |
| CN109327632A (en) * | 2018-11-23 | 2019-02-12 | 深圳前海微众银行股份有限公司 | Intelligent quality inspection system, method and the computer readable storage medium of customer service recording |
| CN110543635A (en) * | 2019-09-05 | 2019-12-06 | 子长科技(北京)有限公司 | information detection method and device based on deep learning and computer storage medium |
| CN110634471A (en) * | 2019-09-21 | 2019-12-31 | 龙马智芯（珠海横琴）科技有限公司 | Voice quality inspection method and device, electronic equipment and storage medium |
| CN110839112A (en) * | 2019-11-18 | 2020-02-25 | 广东电网有限责任公司佛山供电局 | Problem voice detection method and device |
| CN111445121A (en) * | 2020-03-24 | 2020-07-24 | 上海明略人工智能(集团)有限公司 | Risk assessment method and apparatus, storage medium, and electronic apparatus |
| CN111444724A (en) * | 2020-03-23 | 2020-07-24 | 腾讯科技（深圳）有限公司 | Medical question-answer quality testing method and device, computer equipment and storage medium |
-
2020
- 2020-08-06 CN CN202010785633.3A patent/CN111916110B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017041372A1 (en) * | 2015-09-07 | 2017-03-16 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| WO2017076263A1 (en) * | 2015-11-03 | 2017-05-11 | 中兴通讯股份有限公司 | Method and device for integrating knowledge bases, knowledge base management system and storage medium |
| US20180060733A1 (en) * | 2016-08-31 | 2018-03-01 | International Business Machines Corporation | Techniques for assigning confidence scores to relationship entries in a knowledge graph |
| CN109327632A (en) * | 2018-11-23 | 2019-02-12 | 深圳前海微众银行股份有限公司 | Intelligent quality inspection system, method and the computer readable storage medium of customer service recording |
| CN110543635A (en) * | 2019-09-05 | 2019-12-06 | 子长科技(北京)有限公司 | information detection method and device based on deep learning and computer storage medium |
| CN110634471A (en) * | 2019-09-21 | 2019-12-31 | 龙马智芯（珠海横琴）科技有限公司 | Voice quality inspection method and device, electronic equipment and storage medium |
| CN110839112A (en) * | 2019-11-18 | 2020-02-25 | 广东电网有限责任公司佛山供电局 | Problem voice detection method and device |
| CN111444724A (en) * | 2020-03-23 | 2020-07-24 | 腾讯科技（深圳）有限公司 | Medical question-answer quality testing method and device, computer equipment and storage medium |
| CN111445121A (en) * | 2020-03-24 | 2020-07-24 | 上海明略人工智能(集团)有限公司 | Risk assessment method and apparatus, storage medium, and electronic apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 吕克勤;: "客户联络中心的中台能力建设", 信息技术与标准化, no. 07, 10 July 2020 (2020-07-10) * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113128231A (en) * | 2021-04-25 | 2021-07-16 | 深圳市慧择时代科技有限公司 | Data quality inspection method and device, storage medium and electronic equipment |
| CN113128231B (en) * | 2021-04-25 | 2024-09-24 | 深圳市慧择时代科技有限公司 | Data quality inspection method and device, storage medium and electronic equipment |
| CN113345430A (en) * | 2021-06-25 | 2021-09-03 | 上海适享文化传播有限公司 | Multi-field query method based on voice fixed condition |
| CN113345430B (en) * | 2021-06-25 | 2024-05-10 | 上海适享文化传播有限公司 | Multi-field query method based on voice fixed condition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111916110B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109767787B (en) | Emotion recognition method, device and readable storage medium | |
| TWI723528B (en) | Computer-executed event risk assessment method and device, computer-readable storage medium and computing equipment | |
| CN111640436B (en) | Method for providing dynamic customer portraits of conversation objects to agents | |
| CN110147445A (en) | Intension recognizing method, device, equipment and storage medium based on text classification | |
| EP3848797A1 (en) | Automatic parameter value resolution for api evaluation | |
| CN109460455A (en) | A kind of Method for text detection and device | |
| CN111639484A (en) | Method for analyzing seat call content | |
| US10860566B1 (en) | Themes surfacing for communication data analysis | |
| CN111831629B (en) | Data processing method and device | |
| CN106776832B (en) | Processing method, apparatus and system for question and answer interactive log | |
| CN113297365B (en) | User intention judging method, device, equipment and storage medium | |
| CN111739537B (en) | Semantic recognition method and device, storage medium and processor | |
| CN112364664B (en) | Training of intention recognition model, intention recognition method, device and storage medium | |
| CN112256845A (en) | Intention recognition method, device, electronic equipment and computer readable storage medium | |
| CN112907358A (en) | Loan user credit scoring method, loan user credit scoring device, computer equipment and storage medium | |
| WO2022267174A1 (en) | Script text generating method and apparatus, computer device, and storage medium | |
| CN111916110A (en) | Voice quality inspection method and device | |
| CN110880142A (en) | Risk entity acquisition method and device | |
| CN110909126A (en) | Information query method and device | |
| CN111428466A (en) | Legal document analysis method and device | |
| CN112800269A (en) | Conference record generation method and device | |
| KR101478146B1 (en) | Apparatus and method for recognizing speech based on speaker group | |
| CN109558580B (en) | Text analysis method and device | |
| CN111401047A (en) | Method and device for generating dispute focus of legal document and computer equipment | |
| CN116150313A (en) | Data expansion processing method and device |
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 | ||
| PP01 | Preservation of patent right |
Effective date of registration: 20240718 Granted publication date: 20240426 |
|
| PP01 | Preservation of patent right |