CN110265023A - A kind of both sides scene words art real-time online quality detecting method - Google Patents
A kind of both sides scene words art real-time online quality detecting method Download PDFInfo
- Publication number
- CN110265023A CN110265023A CN201910420267.9A CN201910420267A CN110265023A CN 110265023 A CN110265023 A CN 110265023A CN 201910420267 A CN201910420267 A CN 201910420267A CN 110265023 A CN110265023 A CN 110265023A
- Authority
- CN
- China
- Prior art keywords
- data
- art
- quality inspection
- asked questions
- plain text
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000007689 inspection Methods 0.000 claims abstract description 57
- 238000004891 communication Methods 0.000 claims abstract description 26
- 238000001514 detection method Methods 0.000 claims abstract description 15
- 238000012360 testing method Methods 0.000 claims abstract description 11
- 230000001360 synchronised effect Effects 0.000 claims abstract description 8
- 238000011156 evaluation Methods 0.000 claims abstract description 7
- 238000005070 sampling Methods 0.000 claims abstract description 6
- 230000006855 networking Effects 0.000 claims abstract description 4
- 238000012797 qualification Methods 0.000 claims description 30
- 238000013479 data entry Methods 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 230000003416 augmentation Effects 0.000 claims description 6
- 230000005236 sound signal Effects 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 3
- 238000007726 management method Methods 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 4
- 230000003044 adaptive effect Effects 0.000 abstract description 3
- 238000012937 correction Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 7
- 238000012372 quality testing Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue 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
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The present invention relates to a kind of both sides scenes to talk about art real-time online quality detecting method, including hardware networking, setting evaluation and test data, voice input, five steps such as quality inspection operation and sampling observation.One aspect of the present invention data communication system constructs structure simple and flexible, and versatility and extended capability are strong, for talking about the quality inspection of art face-to-face, can effectively meet the needs that all kinds of different phonetic communication occasions used and detected operation；On the other hand in the detection process, voice messaging repeatedly can be appraised through comparison and be checked, speech detection operation identification and evaluation and test precision and efficiency can effectively be improved, and output can be synchronized to testing result in time, convenient for the defects of timely correction voice service work, also update can be synchronized according to practical call scenarios to the technical data for appraising through comparison to voice service in time simultaneously, to further increase the flexibility and adaptive capacity to environment when present invention operation.
Description
Technical field
The present invention relates to a kind of both sides scene words art real-time online quality detecting method, belongs to artificial intelligence, speech communication, natural language
Understand technical field.
Background technique
Information friendship is provided by pure voice mode in fields face-to-face etc. now with telemarketing, phone customer service, sales counter
The mutual and more and more quality in order to improve voice service of related service, it is therefore desirable to offer voice service relevant staff
Call words art quality, many contents such as attitude detected, to meet to providing voice service relevant staff work
The needs examined as quality, for this problem, currently develop it is numerous can be used for speech communication quality testing be
System and technology, if number of patent application is that " 2017101238839 " " a kind of automatic quality detecting method of the voice of calling system and are
The patented technology of system ", but found in the system and technology practical application detected in practice to current various types of voice communication quality,
Complicated, data are constituted and often there is system structure in various degree in the system and technology of these current speech communication quality testings
Processing links are complicated, and the equipment operation difficulty so as to cause voice quality detection operation is big, and using flexible is insufficient, and at data
It is relatively low to manage efficiency, there is certain hysteresis quality, in addition, system and technology that current various types of voice communication quality detects exist
In operation, it is often passively to carry out Data Detection comparison to examine, flexibly retrieval can not be carried out according to working condition is used
It repeatedly checks, the basic data for being checked to voice messaging can not also be carried out in time more according to real work situation
Newly, the voice for being difficult to effectively adapt to complexity in utilization so as to cause the system and technology of current speech communication quality testing leads to
News detection operation needs.
Therefore it is directed to this status, there is an urgent need to develop both sides scenes to talk about art real-time online quality detecting method, to meet reality
The needs used.
Summary of the invention
In order to solve some shortcomings on existing sorting technique, the present invention provides a kind of both sides scene words art real-time online matter
Detecting method talks with real-time quality inspection face-to-face for both sides, and can synchronize output to testing result in time, is convenient for timely correction
The defects of voice service work.
In order to realize effect above-mentioned, a kind of both sides scene words art real-time online quality detecting method is proposed comprising
Following steps:
A kind of both sides scene words art real-time online quality detecting method, includes the following steps:
The first step, hardware networking construct one based on the quality inspection evaluating server based on cloud computing first, and for based on cloud
At least one data entry computer terminal of quality inspection evaluating server dispensing based on calculation and at least one include at least two
Then voice collecting terminal based on parallel with one another and independently operated microphone will be commented based on the quality inspection based on cloud computing
Estimate server and be connected with each other respectively with data entry computer terminal and each voice collecting terminal by communication network, while by base
Quality inspection evaluating server based on cloud computing is that the corresponding data communication protocol of each voice collecting terminal distribution and data are logical
Addressable address and data communication permission are interrogated, is built to complete the words online quality inspection hardware platform of art；
Second step, setting evaluation and test data are first based on cloud by data entry computer terminal after completing first step operation
SQL database is based in quality inspection evaluating server based on calculation, typing words art asked questions data, words art problem reply respectively
Data, and talk with art asked questions data, the independent storage of words art problem answer data progress and establish subdata base, then to base
Text duplicate checking subprogram, audio text conversion subprogram, intelligent language are loaded in quality inspection evaluating server based on cloud computing
Meaning judges that subprogram and qualification rate determine subprogram, and wherein qualification rate determines that subprogram separately sets duplicate checking qualification rate decision content；
Third step, voice input, complete second step operation go forward side by side jargon art quality inspection operation when, by put question to personnel and reply personnel it is same
When operate two parallel with one another and independently operated microphones in same voice collecting terminal, and passed through first by enquirement personnel
Microphone is putd question to, and asked questions audio file is obtained, and is then proposed by microphone to the personnel of enquirement by replying personnel
Problem is replied, and answer problem audio file is obtained, and voice collecting terminal is by the asked questions audio file of acquisition and answer
Problem audio file synchronous transfer to based in the quality inspection evaluating server based on cloud computing in case quality inspection；
4th step, quality inspection operation, receiving voice collecting terminal based on the quality inspection evaluating server based on cloud computing will obtain
Asked questions audio file and after replying problem audio file, audio file is backed up first, while literary by audio
Word conversion routine converts plain text document for asked questions audio file and answer problem audio file respectively, then utilizes
Text duplicate checking subprogram respectively carries out words art asked questions data subdata base in asked questions plain text document and second step
Duplicate checking replies data subdata base to the words art problem in problem plain text document and second step that replies and carries out duplicate checking, then will look into
The qualification rate in second step is delivered to duplicate checking structure after weight to determine in subprogram, determines that subprogram is putd question to current by qualification rate
It problem plain text document duplicate checking rate and replies problem plain text document duplicate checking rate and carries out preliminary judgement, when duplicate checking rate is more than or equal to the
When the duplicate checking qualification rate decision content of two steps setting, then determine that current words art is qualified, and pass result is calculated by data inputting
The output of machine terminal；When duplicate checking rate is less than the duplicate checking qualification rate decision content of second step setting, then sub- journey is judged by the intelligent meaning of one's words
To asked questions plain text document and reply words art asked questions data subdata base, words in problem plain text document and second step
Art problem replies data subdata base and carries out semantics identity comparison, if talking about art asked questions data subdata in the meaning of one's words and second step
Library, the words art problem answer data subdata base meaning of one's words are consistent, then determine that current words art is qualified, and pass result is passed through data record
Enter terminal output, while current asked questions plain text document and answer problem plain text document content augmentation being arrived
Second asked questions data subdata base, words art problem reply it is spare in data subdata base, will be unqualified if unqualified
As a result it is exported by data entry computer terminal, while returning to third step and carrying out voice input again and up to quality inspection qualification is
Only；If duplicate checking rate can not carry out data identification, by current asked questions plain text document and problem plain text document is replied
Directly exported by the data entry computer terminal of the first step, by staff judge current asked questions plain text document and
Whether answer problem plain text document words art is qualified, if qualified, pass result is exported by data entry computer terminal, together
When by current asked questions plain text document and reply problem plain text document content augmentation to second asked questions number
According to subdata base, words art problem reply data subdata base in it is spare, if unqualified, by unqualified result pass through data inputting meter
The output of calculation machine terminal, while returning to third step and carrying out voice input again and until quality inspection is qualified.
The 4th step operation is completed in 5th step, sampling observation, and talk about art detection it is qualified after, from detection it is qualified if in art, pass through
Audio file and with audio file corresponding pure document of the data entry computer terminal from qualified every backup in the first step
File randomly select being inspected by random samples, and Examined is exported by data entry computer terminal, and when sampling observation does not conform to
When lattice, then third step is returned to, acquisition is re-typed to the audio file of present lot.
Further, the voice collecting terminal separately includes audio signal amplifying device and audio signal conversion equipment.
Further, the quality inspection evaluating server based on based on cloud computing, data entry computer terminal and
Voice collecting terminal is connected with each other by gateway and communication network respectively.
Further, in the second step, words art asked questions data subdata base, words art problem reply data subnumber
It is connected with each other according at least one database pointer subprogram is passed through between library, and the database pointer subprogram is separately looked into text
The connection of baryon program.
Further, in the second step, duplicate checking qualification rate decision content be duplicate checking rate 60% -80% be it is qualified, 10% -
60% is unqualified；10% the following are can not identify data
One aspect of the present invention data communication system constructs structure simple and flexible, and versatility and extended capability are strong, for talking about face-to-face
The quality inspection of art can effectively meet the needs that all kinds of different phonetic communication occasions used and detected operation；On the other hand it was detecting
Cheng Zhong repeatedly can be appraised through comparison and be checked to voice messaging, can effectively improve speech detection operation identification and evaluation and test precision and
Efficiency, and output can be synchronized to testing result in time, convenient for the defects of timely correction voice service work, while may be used also
Update is synchronized according to practical call scenarios to the technical data for appraising through comparison to voice service in time, to further increase
Flexibility and adaptive capacity to environment when the present invention is run.
Detailed description of the invention
The following describes the present invention in detail with reference to the accompanying drawings and specific embodiments；
Fig. 1 is the method for the present invention flow chart.
Fig. 2 is that the present invention executes flow diagram；
Fig. 3 is that present system constitutes structural schematic diagram.
Specific embodiment
To be easy to understand the technical means, the creative features, the aims and the efficiencies achieved by the present invention, below with reference to
Specific embodiment, the present invention is further explained.
Art real-time online quality detecting method is talked about at a kind of both sides scene as described in Fig. 1-3, is included the following steps:
The first step, hardware networking construct one based on the quality inspection evaluating server based on cloud computing first, and for based on cloud
At least one data entry computer terminal of quality inspection evaluating server dispensing based on calculation and at least one include at least two
Then voice collecting terminal based on parallel with one another and independently operated microphone will be commented based on the quality inspection based on cloud computing
Estimate server and be connected with each other respectively with data entry computer terminal and each voice collecting terminal by communication network, while by base
Quality inspection evaluating server based on cloud computing is that the corresponding data communication protocol of each voice collecting terminal distribution and data are logical
Addressable address and data communication permission are interrogated, is built to complete the words online quality inspection hardware platform of art；
Second step, setting evaluation and test data are first based on cloud by data entry computer terminal after completing first step operation
SQL database is based in quality inspection evaluating server based on calculation, typing words art asked questions data, words art problem reply respectively
Data, and talk with art asked questions data, the independent storage of words art problem answer data progress and establish subdata base, then to base
Text duplicate checking subprogram, audio text conversion subprogram, intelligent language are loaded in quality inspection evaluating server based on cloud computing
Meaning judges that subprogram and qualification rate determine subprogram, and wherein qualification rate determines that subprogram separately sets duplicate checking qualification rate decision content；
Third step, voice input, complete second step operation go forward side by side jargon art quality inspection operation when, by put question to personnel and reply personnel it is same
When operate two parallel with one another and independently operated microphones in same voice collecting terminal, and passed through first by enquirement personnel
Microphone is putd question to, and asked questions audio file is obtained, and is then proposed by microphone to the personnel of enquirement by replying personnel
Problem is replied, and answer problem audio file is obtained, and voice collecting terminal is by the asked questions audio file of acquisition and answer
Problem audio file synchronous transfer to based in the quality inspection evaluating server based on cloud computing in case quality inspection；
4th step, quality inspection operation, receiving voice collecting terminal based on the quality inspection evaluating server based on cloud computing will obtain
Asked questions audio file and after replying problem audio file, audio file is backed up first, while literary by audio
Word conversion routine converts plain text document for asked questions audio file and answer problem audio file respectively, then utilizes
Text duplicate checking subprogram respectively carries out words art asked questions data subdata base in asked questions plain text document and second step
Duplicate checking replies data subdata base to the words art problem in problem plain text document and second step that replies and carries out duplicate checking, then will look into
The qualification rate in second step is delivered to duplicate checking structure after weight to determine in subprogram, determines that subprogram is putd question to current by qualification rate
It problem plain text document duplicate checking rate and replies problem plain text document duplicate checking rate and carries out preliminary judgement, when duplicate checking rate is more than or equal to the
When the duplicate checking qualification rate decision content of two steps setting, then determine that current words art is qualified, and pass result is calculated by data inputting
The output of machine terminal；When duplicate checking rate is less than the duplicate checking qualification rate decision content of second step setting, then sub- journey is judged by the intelligent meaning of one's words
To asked questions plain text document and reply words art asked questions data subdata base, words in problem plain text document and second step
Art problem replies data subdata base and carries out semantics identity comparison, if talking about art asked questions data subdata in the meaning of one's words and second step
Library, the words art problem answer data subdata base meaning of one's words are consistent, then determine that current words art is qualified, and pass result is passed through data record
Enter terminal output, while current asked questions plain text document and answer problem plain text document content augmentation being arrived
Second asked questions data subdata base, words art problem reply it is spare in data subdata base, will be unqualified if unqualified
As a result it is exported by data entry computer terminal, while returning to third step and carrying out voice input again and up to quality inspection qualification is
Only；If duplicate checking rate can not carry out data identification, by current asked questions plain text document and problem plain text document is replied
Directly exported by the data entry computer terminal of the first step, by staff judge current asked questions plain text document and
Whether answer problem plain text document words art is qualified, if qualified, pass result is exported by data entry computer terminal, together
When by current asked questions plain text document and reply problem plain text document content augmentation to second asked questions number
According to subdata base, words art problem reply data subdata base in it is spare, if unqualified, by unqualified result pass through data inputting meter
The output of calculation machine terminal, while returning to third step and carrying out voice input again and until quality inspection is qualified.
The 4th step operation is completed in 5th step, sampling observation, and talk about art detection it is qualified after, from detection it is qualified if in art, pass through
Audio file and with audio file corresponding pure document of the data entry computer terminal from qualified every backup in the first step
File randomly select being inspected by random samples, and Examined is exported by data entry computer terminal, and when sampling observation does not conform to
When lattice, then third step is returned to, acquisition is re-typed to the audio file of present lot.
In the present embodiment, the voice collecting terminal separately includes audio signal amplifying device and audio signal converting means
It sets, and the quality inspection evaluating server based on based on cloud computing, data entry computer terminal and voice collecting terminal
It is connected with each other respectively by gateway and communication network.
In addition, words art asked questions data subdata base, words art problem reply data subdata base in the second step
Between be connected with each other by least one database pointer subprogram, and the database pointer subprogram separately with text duplicate checking
Program connection.
It needing to stress, in the second step, duplicate checking qualification rate decision content is that duplicate checking rate 60% -80% is qualification,
10% -60% is unqualified；10% the following are can not identify data.
The present invention is in specific implementation:
Voice collecting terminal each first may be interspersed within any position within the scope of larger job, and each wheat of same voice collecting terminal
Gram wind also may be interspersed within any position within the scope of larger job, pass through communication network between each microphone of same voice collecting terminal
Network constitutes a communication local area network；
Secondly, when carrying out quality inspection operation, when passing through third step voice input, the audio number of some voice collecting terminal acquisition
The completely new data being not present in data subdata base are replied according to containing words art asked questions data subdata base, talking about art problem,
It is cached in quality inspection evaluating server first to the completely new data based on based on cloud computing, and works as multiple voice collectings
When the completely new data occurs in terminal, the completely new data amended record to words art asked questions data subdata base, words art problem are answered
In complex data subdata base；When same voice collecting terminal is in special time period, such as in 24 hours, when specific in 72 hours
Between frequently occur in the period, then the completely new data amended record to words art asked questions data subdata base, words art problem are replied into number
According in subdata base, to complete dialogue art asked questions data subdata base, words art problem replies number in data subdata base
According to active update is carried out, the adaptability when present invention is run to use occasion environment is improved.
Finally, the present invention when carrying out quality inspection operation, passes through duplicate checking qualification rate decision content, directly by data to be tested first
It is divided into qualified data, unqualified data and unidentified data, by carrying out classification processing to data, effectively improves data and directly make
Industry efficiency, and unqualified data and unidentified data then further identify quality inspection by means such as subsequent artefacts' detections, to mention
The precision of high quality inspection operation, while data after quality inspection qualification need to separately be inspected by random samples, to further improve voice data matter
The accuracy of inspection.
According to above-described embodiment, one aspect of the present invention data communication system constructs structure simple and flexible, versatility and extension
Ability is strong, and for the quality inspection for the middle words art that talks face to face, can effectively meet all kinds of different phonetics communication occasions use and
Detect the needs of operation；On the other hand in the detection process, voice messaging repeatedly can be appraised through comparison and is checked, can effectively be mentioned
High speech detection operation identification and evaluation and test precision and efficiency, and output can be synchronized to testing result in time, convenient for entangling in time
The defects of positive voice service work, while can also be in time to the technical data for being appraised through comparison to voice service according to practical call
Situation synchronizes update, to further increase the flexibility and adaptive capacity to environment when present invention operation.
The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (5)
1. art real-time online quality detecting method is talked about at a kind of both sides scene, which is characterized in that the mobile terminal and intercomputer is logical
News management method includes the following steps:
The first step, hardware networking construct one based on the quality inspection evaluating server based on cloud computing first, and for based on cloud
At least one data entry computer terminal of quality inspection evaluating server dispensing based on calculation and at least one include at least two
Then voice collecting terminal based on parallel with one another and independently operated microphone will be commented based on the quality inspection based on cloud computing
Estimate server and be connected with each other respectively with data entry computer terminal and each voice collecting terminal by communication network, while by base
Quality inspection evaluating server based on cloud computing is that the corresponding data communication protocol of each voice collecting terminal distribution and data are logical
Addressable address and data communication permission are interrogated, is built to complete the words online quality inspection hardware platform of art；
Second step, setting evaluation and test data are first based on cloud by data entry computer terminal after completing first step operation
SQL database is based in quality inspection evaluating server based on calculation, typing words art asked questions data, words art problem reply respectively
Data, and talk with art asked questions data, the independent storage of words art problem answer data progress and establish subdata base, then to base
Text duplicate checking subprogram, audio text conversion subprogram, intelligent language are loaded in quality inspection evaluating server based on cloud computing
Meaning judges that subprogram and qualification rate determine subprogram, and wherein qualification rate determines that subprogram separately sets duplicate checking qualification rate decision content；
Third step, voice input, complete second step operation go forward side by side jargon art quality inspection operation when, by put question to personnel and reply personnel it is same
When operate two parallel with one another and independently operated microphones in same voice collecting terminal, and passed through first by enquirement personnel
Microphone is putd question to, and asked questions audio file is obtained, and is then proposed by microphone to the personnel of enquirement by replying personnel
Problem is replied, and answer problem audio file is obtained, and voice collecting terminal is by the asked questions audio file of acquisition and answer
Problem audio file synchronous transfer to based in the quality inspection evaluating server based on cloud computing in case quality inspection；
4th step, quality inspection operation, receiving voice collecting terminal based on the quality inspection evaluating server based on cloud computing will obtain
Asked questions audio file and after replying problem audio file, audio file is backed up first, while literary by audio
Word conversion routine converts plain text document for asked questions audio file and answer problem audio file respectively, then utilizes
Text duplicate checking subprogram respectively carries out words art asked questions data subdata base in asked questions plain text document and second step
Duplicate checking replies data subdata base to the words art problem in problem plain text document and second step that replies and carries out duplicate checking, then will look into
The qualification rate in second step is delivered to duplicate checking structure after weight to determine in subprogram, determines that subprogram is putd question to current by qualification rate
Problem plain text document duplicate checking rate and answer problem plain text document duplicate checking rate carry out preliminary judgement,
When duplicate checking rate is more than or equal to the duplicate checking qualification rate decision content of second step setting, then determine that current words art is qualified, and will close
Lattice result is exported by data entry computer terminal；
When duplicate checking rate is less than the duplicate checking qualification rate decision content of second step setting, then judge that sub- journey asks enquirement by the intelligence meaning of one's words
It inscribes plain text document and replies problem plain text document and talk about art asked questions data subdata base in second step, words art problem is answered
Complex data subdata base carries out semantics identity comparison, if talking about art asked questions data subdata base, words art in the meaning of one's words and second step
The problem answer data subdata base meaning of one's words is consistent, then determines that current words art is qualified, and pass result is passed through data inputting calculating
The output of machine terminal, while by current asked questions plain text document and problem plain text document content augmentation is replied to second
Asked questions data subdata base, words art problem reply data subdata base in it is spare, if unqualified, unqualified result is led to
The output of data entry computer terminal is crossed, while returning to third step and carrying out voice input again and until quality inspection is qualified；
If duplicate checking rate can not carry out data identification, by current asked questions plain text document and problem plain text document is replied
Directly exported by the data entry computer terminal of the first step, by staff judge current asked questions plain text document and
Whether answer problem plain text document words art is qualified, if qualified, pass result is exported by data entry computer terminal, together
When by current asked questions plain text document and reply problem plain text document content augmentation to second asked questions number
According to subdata base, words art problem reply data subdata base in it is spare, if unqualified, by unqualified result pass through data inputting meter
The output of calculation machine terminal, while returning to third step and carrying out voice input again and until quality inspection is qualified；
The 4th step operation is completed in 5th step, sampling observation, and talk about art detection it is qualified after, from detection it is qualified if in art, pass through first
Audio file and with audio file corresponding pure document files of the data entry computer terminal from qualified every backup in step
Randomly select being inspected by random samples, and Examined exported by data entry computer terminal, and when inspecting unqualified by random samples,
Third step is then returned, acquisition is re-typed to the audio file of present lot.
2. art real-time online quality detecting method is talked about at a kind of both sides scene according to claim 1, it is characterised in that: the language
Sound acquisition terminal separately includes audio signal amplifying device and audio signal conversion equipment.
3. art real-time online quality detecting method is talked about at a kind of both sides scene according to claim 1, it is characterised in that: the base
Quality inspection evaluating server, data entry computer terminal and voice collecting terminal based on cloud computing pass through gateway respectively
It is connected with each other with communication network.
4. art real-time online quality detecting method is talked about at a kind of both sides scene according to claim 1, it is characterised in that: described the
In two steps, words art asked questions data subdata base, words art problem reply and pass through at least one data between data subdata base
Library pointer subprogram is connected with each other, and the database pointer subprogram is separately connect with text duplicate checking subprogram.
5. art real-time online quality detecting method is talked about at a kind of both sides scene according to claim 1, it is characterised in that: described the
In two steps, duplicate checking qualification rate decision content is that duplicate checking rate 60% -80% is qualification, and 10% -60% is unqualified；10% the following are can not
Identify data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910420267.9A CN110265023A (en) | 2019-05-20 | 2019-05-20 | A kind of both sides scene words art real-time online quality detecting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910420267.9A CN110265023A (en) | 2019-05-20 | 2019-05-20 | A kind of both sides scene words art real-time online quality detecting method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110265023A true CN110265023A (en) | 2019-09-20 |
Family
ID=67914827
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910420267.9A Pending CN110265023A (en) | 2019-05-20 | 2019-05-20 | A kind of both sides scene words art real-time online quality detecting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110265023A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101572756A (en) * | 2008-04-29 | 2009-11-04 | 台达电子工业股份有限公司 | Dialogue system and voice dialogue processing method |
| CN105975514A (en) * | 2016-04-28 | 2016-09-28 | 朱宇光 | Automatic quality testing method and system |
| CN106934000A (en) * | 2017-03-03 | 2017-07-07 | 深圳市彬讯科技有限公司 | A kind of automatic quality detecting method of the voice of calling system and system |
| CN107317942A (en) * | 2017-07-18 | 2017-11-03 | 国家电网公司客户服务中心南方分中心 | A kind of call center's customer service system is recognized and monitoring system with online voice mood |
| US10129392B1 (en) * | 2017-08-25 | 2018-11-13 | Global Tel*Link Corporation | Systems and methods for detecting inmate to inmate conference calls |
| CN109446524A (en) * | 2018-10-25 | 2019-03-08 | 第四范式（北京）技术有限公司 | A kind of voice quality detecting method and device |
| CN109740155A (en) * | 2018-12-27 | 2019-05-10 | 广州云趣信息科技有限公司 | A kind of customer service system artificial intelligence quality inspection rule self concludes the method and system of model |
-
2019
- 2019-05-20 CN CN201910420267.9A patent/CN110265023A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101572756A (en) * | 2008-04-29 | 2009-11-04 | 台达电子工业股份有限公司 | Dialogue system and voice dialogue processing method |
| CN105975514A (en) * | 2016-04-28 | 2016-09-28 | 朱宇光 | Automatic quality testing method and system |
| CN106934000A (en) * | 2017-03-03 | 2017-07-07 | 深圳市彬讯科技有限公司 | A kind of automatic quality detecting method of the voice of calling system and system |
| CN107317942A (en) * | 2017-07-18 | 2017-11-03 | 国家电网公司客户服务中心南方分中心 | A kind of call center's customer service system is recognized and monitoring system with online voice mood |
| US10129392B1 (en) * | 2017-08-25 | 2018-11-13 | Global Tel*Link Corporation | Systems and methods for detecting inmate to inmate conference calls |
| CN109446524A (en) * | 2018-10-25 | 2019-03-08 | 第四范式（北京）技术有限公司 | A kind of voice quality detecting method and device |
| CN109740155A (en) * | 2018-12-27 | 2019-05-10 | 广州云趣信息科技有限公司 | A kind of customer service system artificial intelligence quality inspection rule self concludes the method and system of model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112966082B (en) | Audio quality inspection method, device, equipment and storage medium | |
| CN105141787A (en) | Service record compliance checking method and device | |
| CN108270640B (en) | Information interoperation consistency test method for intelligent marketing and distribution information integration system | |
| CN111061696B (en) | Method and device for analyzing transaction message log | |
| CN110929011A (en) | Conversation analysis method, device and equipment | |
| CN109756395A (en) | A kind of business datum monitoring method and system | |
| CN112734171A (en) | Technical service integrated management system | |
| CN112511696A (en) | System and method for identifying bad content of call center AI engine | |
| CN110275509B (en) | Energy storage power station monitoring function test method and system | |
| CN117519951B (en) | Real-time data processing method and system based on message center | |
| CN113592305A (en) | Test method, test device, electronic device, and storage medium | |
| CN110265023A (en) | A kind of both sides scene words art real-time online quality detecting method | |
| CN206039610U (en) | Join in marriage net electric power engineering and check and accept system | |
| CN112990746A (en) | Cloud platform intelligent management system of inspection and detection mechanism | |
| CN116828109A (en) | Intelligent evaluation method and system for telephone customer service quality | |
| CN103345466A (en) | Academic paper information detection method based on free internet information | |
| CN116628061A (en) | Multi-source heterogeneous data acquisition method, device, equipment and storage medium | |
| CN116385137A (en) | Enterprise anti-fraud risk assessment method and system based on power information data | |
| CN112446619B (en) | Power distribution network rush-repair processing method and device | |
| CN115641003A (en) | Method for hierarchical management and control of associated risks and troubleshooting and treatment of hidden dangers | |
| CN115293750A (en) | AI (AI) middle desk-based intelligent auditing system, method and device | |
| CN109618065A (en) | A kind of voice quality inspection rating system | |
| CN111965995B (en) | Three-in-one aircraft avionics system health assessment and prediction system and method thereof | |
| CN103646439A (en) | Routing inspection method and routing inspection system with conversation function | |
| CN114564405A (en) | Test case checking method and system based on log monitoring |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190920 |