CN106803797A - A kind of VoIP speech quality method for objectively evaluating based on network parameter - Google Patents
A kind of VoIP speech quality method for objectively evaluating based on network parameter Download PDFInfo
- Publication number
- CN106803797A CN106803797A CN201710126997.9A CN201710126997A CN106803797A CN 106803797 A CN106803797 A CN 106803797A CN 201710126997 A CN201710126997 A CN 201710126997A CN 106803797 A CN106803797 A CN 106803797A
- Authority
- CN
- China
- Prior art keywords
- network parameter
- voip
- network
- speech
- data
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5003—Managing SLA; Interaction between SLA and QoS
- H04L41/5009—Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0823—Errors, e.g. transmission errors
- H04L43/0829—Packet loss
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
- H04L43/087—Jitter
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Telephonic Communication Services (AREA)
Abstract
A kind of VoIP speech quality method for objectively evaluating based on network parameter of this paper disclosure of the invention, belongs to voice quality assessment research field.This method is obtained to the network parameter in VoIP communication process first, and then the network parameter for collecting is pre-processed, including average, the basic statistics amount based on variance, most value, mode and median；Wherein network parameter refers mainly to packet loss, jitter value and delay value；PCA dimension-reduction treatment then is carried out to pretreated data, finally by the data input after dimensionality reduction to multiple linear regression model, speech quality objective assessment score is calculated, the evaluation to VoIP speech qualities based on network parameter is thus achieved.The present invention proposes a kind of new VoIP speech quality method for objectively evaluating, it is not necessary to obtain voice signal, it is only necessary to which monitor network parameter values just can make accurately evaluation with voice quality.
Description
Technical field
It is that one kind is suitable in real time the present invention relates to a kind of VoIP speech quality method for objectively evaluating based on network parameter
VoIP speech quality objective assessment methods under call scene, belong to voice quality assessment technical field.
Background technology
With the fast development of Internet technology, IP phone is widely used by its huge price advantage.
It is well known that the transmission criterion of IP network is " doing one's best ", this also implies that he can not possibly be as those are exclusively for real-time
Network designed by voice communication has reliability very high like that.(such as packet loss, trembles the network characterization that IP network has
It is dynamic, time delay) understand the speech quality largely influenceed in communication process.Therefore, how accurately to VoIP speech qualities
Evaluation is made just to be particularly important.The pioneer of the convergence service network that VoIP is transmitted as the next generation based on packet, as it
The measurement of the voice quality of main business will provide reference and experience for the qos measurement of future IP network.
At present, the evaluation method of voice quality is divided into subjective assessment and the class of objective evaluation two.Subjective evaluation method can be accurate
Voice quality makes evaluation, mean opinion score (MOS, the Mean for generally P.800 and P.830 being advised using ITU-T
Opinion Score) method of testing, wherein widely used method of testing be absolute classification deciding grade and level test (ACR,
Absolute Category Rating).But subjective assessment operation is relative complex, it is necessary to substantial amounts of manpower and materials support, uncomfortable
Share in real-time network assessment.
In view of many difficulties of subjective assessment, wish to use conveniently objective means voice quality in practical application
Evaluated.PESQ methods are a kind of objective speech quality assessment methods of the ITU-T in P.862 middle proposition, are also application at present
A kind of most commonly used method, it is a kind of typical evaluation method based on input-output, can be used for test and is arrived by end
Hold the quality of the voice of transmission.But PESQ cannot shake to packet loss, and the IP network factor based on time delay makes measurement, 2011
Year ITU-T develops POLQA again, and its main feature can cover newest voice coding and network transmission technology, for 3G,
4G/LTE, with accuracy higher during voip network.But POLQA complete equipments are costly, and needed simultaneously as PESQ
The speech of row input transmitting terminal and receiving terminal, is not suitable for the monitoring of real-time IP network speech quality and IP network parameter
Optimization.
Objective examination is carried out to VoIP speeches mainly given a mark using PESQ algorithms, evaluated as described above, existing, this
Plant evaluation method and be primarily present two problems for VoIP speeches：One be for network delay present in VoIP speeches, shake,
Packet loss problem cannot make accurately assessment, and then influence the accuracy of whole test result.Meanwhile, by contrasting subjective testing
As a result, also confirm that the degree of accuracy is not high when being evaluated VoIP speeches using PESQ.Two is that PESQ is a kind of commenting for input-output
Valency method then carries out comparative evaluation, it is necessary to collect original input voice and by the output voice after VoIP system, this
Sample cannot be just met in real time phone call scene, and VoIP speech qualities are monitored.Therefore, in order to further lift objective calculation
Method is to the accuracy of VoIP Speech Assessments, and can obtain the speech quality under real time phone call state, and this is accomplished by by profit
The evaluation of non-insertion (without reference) is realized to VoIP speeches with many network parameters under current IP transmission networks.In the world
ITU-T G.107 standard pin to VoIP speech quality Test Suggestions for the computation model E- models (E- of transmission plan
Model), the model is evaluated and tested based on traditional voice code encoding/decoding mode and network parameter, but for current mobile Internet
The voice quality assessment degree of accuracy of middle VoIP calls software is not good enough.China authorizes Publication No. CN105959453A (publication date：
On September 21st, 2016) the patent test system and its method of testing of VOIP " multi-channel synchronous test ", disclose and a kind of lead to more
The test system of road synchronism detection VOIP, VOIP server unit has multiple voice channels, for testing the voice channel
And the VOIP equipment described to be measured being connected with the voice channel, so as to test the connectedness of VoIP speech channels, do not consider
The speech quality of call.
The content of the invention
Cannot make accurate and real-time to VoIP speech qualities to solve existing speech method for evaluating objective quality
Evaluate, and existing survey tool there is a problem of it is expensive, need input original speech refer to based on, the present invention public affairs
A kind of VoIP voice quality objective evaluation methods based on network parameter are opened.
To achieve these goals, the basic ideas of the inventive method are：(lost by IP network parameter according to VoIP speeches
Bag, shake, time delay) the characteristics of influence huge, by using this major influence factors of network parameter, with reference to PCA, regression analysis
Based on technological means, in the case where that need not be input into and export voice, only VoIP speeches are entered by obtaining network parameter
Row objective evaluation, is specifically evaluated VoIP speech qualities in real time using network parameter, is obtained under real-time VoIP calls
Network parameter, then carries out basic data prediction and dimensionality reduction, finally the data after dimensionality reduction to the network parameter for obtaining
It is input in the multiple linear regression model obtained by training, obtains objective assessment score, so as to reacts VoIP in current network
Speech quality under state.
A kind of VoIP speech quality method for objectively evaluating based on network parameter, in the case of being conversed in real-time network
Network parameter in VoIP communication process is acquired, data processing and drop are then carried out to the network parameter for collecting
Operation based on dimension, finally obtains VoIP speech quality objective assessment scores to the data separate Mathematical Modeling after treatment.
A kind of VoIP voice quality objective evaluation methods based on network parameter, comprise the following steps：
Step one, during real time phone call voip network parameter is acquired, obtains many set of network parameters；
Step one is specially：In the T seconds interval in a communication process primary network ginseng is obtained every t seconds (T >=t)
Number, obtains T/t set of network parameters, and packet loss, jitter value and delay value are included per set of network parameters；
Step 2, many set of network parameters to being collected in step one are pre-processed, and obtain corresponding multiple pretreatments
Data；
Step 2 is specially：Ask for average, variance, most value, mode and the middle position of multi-group data under every kind of network parameter
Basic mathematical statistic based on number, obtains N number of preprocessed data { X1, X2..., Xn, n=1,2 ..., N, N are positive integer }；
Step 3, the preprocessed data obtained to step 2 carry out dimension-reduction treatment using PCA principal component analytical methods, obtain
Low-dimensional data；
Step 3, specifically：Pretreated high dimensional data is converted to the low-dimensional data { Y for representing principal component1,
Y2..., Ym, m=1,2 ..., M, M are positive integer, and M≤N }；
Step 4, to being calculated by the low-dimensional data obtained in step 3, obtain VoIP speech qualities objective evaluation point
Number；
Step 4, specially：The low-dimensional data that step 3 is exported is calculated using multiple linear regression model, is obtained
VoIP speech quality objective assessment score S, i.e. S=β in T seconds communication process0+β1Y1+β2Y2+…+βmYm, factor beta therein0,
β1, β2..., βmObtained by precondition, its step is as follows：
Step 4.1 collection P (P >=200) bar time interval is the VoIP real time phone call speeches of T seconds, obtains a plurality of speech letter
Number；And be acquired according to described in step one simultaneously, obtain the corresponding network parameter of every speech；
Every voice signal that step 4.2 is collected to step 4.1 again, using subjective evaluation method (such as ITU-T
P.800 the MOS subjective evaluation methods advised) or method for objectively evaluating (POLQA that P.863 such as ITU-T advises is objective to be commented
Valency method), obtain P speech quality fraction (i.e. MOS points)；Meanwhile, to the P set of network parameters that collects according to step 2 and three
Processed, the low-dimensional data and speech quality MOS obtained using step 3 point is trained by the method for multiple linear regression
To the multiple linear regression model for VoIP speech quality objective evaluations, that is, the instruction in objective assessment score S computing formula
Practice coefficient；
Wherein, the method for described multiple linear regression can use least square method；
So far, from step one to step 4, a kind of VoIP voice qualities objective evaluation side based on network parameter is completed
Method.
Beneficial effect
A kind of VoIP voice quality objective evaluation methods based on network parameter of the present invention, contrast existing skill
Art, has the advantages that：
1. the method for the invention can be simple to utilize network parameter pair in the case where input, output voice is not obtained
VoIP speech qualities are evaluated, i.e., the IP network speech objective evaluation based on multiple network parameters, improve existing speech visitor
Seeing evaluation criterion PESQ (ITU-T is P.862) cannot make the shortcoming of accurate and Real-Time Evaluation to VoIP speeches；
2. the network parameter that the method for the invention is based on obtaining in real time is input into as algorithm, to the network of Real-time Collection
After parameter pretreatment, using PCA dimensionality reductions, objective assessment score finally is obtained using multiple linear regression model, that is, avoid master
A large amount of human and material resources needed for observing examination, while also can approximately replace to POLQA costly, that is, overcome newest standards
POLQA (ITU-T is P.863) is expensive and the deficiency that needs input speech signal to measure；
3. being trained in the method for the invention and testing used language material, data carries out real time phone call by VoIP softwares
Obtain；The accuracy of VoIP speech quality objective evaluations is improved to a certain extent, while simplifying testing process, is realized
Relatively accurate to VoIP speeches and real-time objective evaluation.
Brief description of the drawings
A kind of operating process schematic diagram of the VoIP voice quality objective evaluation methods based on network parameter of Fig. 1 present invention；
Training in a kind of VoIP voice quality objective evaluation embodiments of the method based on network parameter of Fig. 2 present invention is used
Language material gathers schematic diagram；
Multiple regression in a kind of VoIP voice quality objective evaluation embodiments of the method based on network parameter of Fig. 3 present invention
The scatter diagram of analysis method data result.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.
Embodiment 1
The present embodiment describes the operating process based on the method for the invention, as shown in Figure 1.
It will be seen from figure 1 that the operating process of the present invention and embodiment is as follows：
Step 1：Connection VoIP softwares are conversed and are gathered network parameter, in corresponding diagram 11；
By Fig. 1 it can also be seen that the present invention and the network parameter of the present embodiment collection are mainly packet loss, jitter value and prolong
Duration；
The collection of network parameter is using the VoIP call softwares based on webRTC designs in the present embodiment.Such VoIP calls
Software can carry out Real-time Collection network parameter and voice signal.
In the present embodiment, network parameter is recorded once each second in a communication process.In order to test heterogeneous networks
The accuracy that VoIP speeches under environment are evaluated, during network parameter is gathered, using network harm simulation software (WANem)
Packet loss, shake, delay parameter setting are carried out, traversal is by good to poor various network environments.Every time by two in VoIP communication process
Name tester reads aloud the voice signal of the language material text of standard, collection transmitting terminal and receiving terminal, and by VoIP software synchronizations
Obtain network parameter.
For the situation of more real reflection network, in the present embodiment, while gathering packet loss, time delay, jitter parameter
And voice signal, during assessment algorithm is modeled, test network parameter used is using the collection of real network real time phone call
Method.
Step 2：Network parameter to being obtained by step 1 carries out data prediction, including average, variance, most value, mode
With the basic statistics amount such as median, and pretreated data are carried out with dimensionality reduction, in corresponding diagram 12；
In the present embodiment, average level of the voice quality grade of a certain period not only with now network parameter is relevant,
Fluctuation also with network parameter is relevant.In order to reflect the fluctuation situation of network parameter, we carried out 12 network ginsengs in 12 seconds
Number is extracted, and calculates the statistic based on the average of 12 set of network parameters observations, variance, most value.Specifically, if i-th 12
The observation matrix that obtains of second is：
Wherein：
PLij--- represent i-th j-th packet loss parameter of segment of speech；
Dij--- represent i-th j-th delay parameter of segment of speech；
Jij--- represent i-th j-th jitter parameter of segment of speech.
Network parameter observation to collecting is pre-processed, including to ask for current speech sample respectively corresponding 12 groups
The basic statistics amounts such as the average of network parameter (packet loss, time delay and shake), variance, maximum, minimum value, median and mode,
Obtain [Xi1, Xi2 ..., Xi18], wherein, XijIt is the statistic of PL, D, J, [Xi1, Xi2..., Xi6] it is the statistic of packet loss,
[Xi7, Xi8 ..., Xi12] it is the statistic of time delay, [Xi13, Xi14..., Xi18] it is the statistic shaken.
Step 3：Pretreated data to being obtained by step 2 carry out Principle component extraction, second of 2 in corresponding diagram 1
Part；
In the present embodiment, data dimension is larger after pretreatment, relation between hyperspace analyze data can difficulty compared with
Greatly, and between each parameter there is stronger correlation, the influence for intersecting is produced to VoIP voice qualities, to analyze and state this
Plant cross influence relatively difficult.Therefore, introducing PCA.Specifically, in analytical procedure 1 between matrix p column elements
Correlation, its principal component is calculated using principal component analysis (Principal Component Analysis, PCA), takes preceding 4 sides
Input vector of the larger principal component as regression analysis is differed from, i.e.,：
Wherein：
Ypi--- represent p-th i-th principal component of segment of speech.
Step 4：To the low-dimensional data after the dimensionality reduction for obtaining by step 3, calculate objective using multiple linear regression model
Fraction, voice quality makes corresponding evaluation, in corresponding diagram 13.
In the present embodiment, believe, it is necessary to gather a plurality of speech in advance to obtain the mapping relations of multiple linear regression model
Number and corresponding network parameter be trained, collecting flowchart is as shown in Figure 2.
Figure it is seen that the gatherer process of the present invention and data and speech needed for embodiment is as follows：
Step 1：Data are built in Mute Room 1 and damages unit with controlling network in speech acquisition system, in such as Fig. 2
The parts of Mute Room 1；
In the present embodiment, the lower accuracy evaluated is damaged in order to reflect heterogeneous networks, network is utilized during collection
Damage simulation software (WANem) adds certain packet loss, time delay and shake, in quiet room Mute Room 1, by notebook
After computer connection route device access network, network signal is outwards distributed using specific software.Mobile phone A connection is sent by notebook
Network signal, realize IP call in network control purpose.
After mobile phone A access network, VoIP softwares are connected, call request is sent to the mobile phone B in Mute Room 2.
Step 2：In Mute Room 2 IP calls, the parts of Mute Room 2 in such as Fig. 2 are connected using another mobile phone；
In the present embodiment, the mobile phone B in quiet room Mute Room 2 is by after router access network, connecting
VoIP softwares, the call request to mobile phone A is responded, and realizes that IP converses.
In the present embodiment, the good and bad situation of network can be carried out by the Network Simulation Software in step 1 in communication process
Regulation, obtains the different voice signals and network parameter for damaging quality.
Regression analysis can exactly measure the height of the degree of correlation and regression fit degree between each factor, improve
The effect of predictive equation formula.
Data after PCA are carried out into multiple linear regression analysis, polynary assessment formula is obtained, its citation form is：
S=β0+β1Y1+β2Y2+…+βmYm
Wherein Y1, Y2..., YmIt is the principal component for extracting, βiIt is fitting coefficient.The final form of objective assessment score S and
Coefficient is obtained by Multivariable regressive analysis model, such as following formula：
S=-0.0807*Y1+0.0466*Y2+0.1036*Y3+0.1612*Y4+2.4671
Embodiment 2
The present embodiment describes the objective evaluation result based on the method for the invention.
As can be seen from Figure 3 the present invention and embodiment it is objective predict the outcome with actual POLQA obtain evaluation result it
Between scatter diagram.
In the present invention and embodiment, evaluated predicting the outcome using VoIP measured datas：Join for every group of network
Number predicts objective score value using this algorithm, and is contrasted with the MOS values (POLQA assessment results) of actual speech, and Fig. 3 is this algorithm
The scatter diagram that predicts the outcome, abscissa predicts the outcome for evaluation model, i.e. the prediction fraction of present invention model used；Ordinate
It is the objective MOS fractions of actual measurement that POLQA is obtained, middle oblique line is isopleth, represents that predicted value is equal with actual value, upper and lower two
Line is represented and point differs from 0.5.As can be seen that the uniformity of predicted value of the invention and actual value is higher from figure, can accurately comment
Valency VoIP speech qualities.
Above-described specific descriptions, purpose, technical scheme and beneficial effect to inventing have been carried out further specifically
It is bright, should be understood that and the foregoing is only specific embodiment of the invention, the protection model being not intended to limit the present invention
Enclose, all any modification, equivalent substitution and improvements within the spirit and principles in the present invention, done etc. should be included in the present invention
Protection domain within.
Claims (9)
1. a kind of VoIP voice quality objective evaluation methods based on network parameter, it is characterised in that：Basic ideas are：According to
The characteristics of VoIP speeches are influenceed huge by IP network parameter (packet loss, shake, time delay), this main shadow by using network parameter
The factor of sound, with reference to the technological means based on PCA, regression analysis, in the case where that need not be input into and export voice, only by obtaining
Take network parameter carries out objective evaluation to VoIP speeches, and VoIP speech qualities are commented in real time using network parameter specifically
Valency, obtains the network parameter under real-time VoIP call, then the network parameter for obtaining is carried out basic data prediction and
Dimensionality reduction, finally in the data input after dimensionality reduction to the multiple linear regression model obtained by training, obtains objective assessment score,
So as to react speech qualities of the VoIP under current network state；It is logical to VoIP in the case of especially by being conversed in real-time network
Network parameter during words is acquired, and then the network parameter for collecting is carried out based on data processing and dimensionality reduction
Operation, finally obtains VoIP speech quality objective assessment scores to the data separate Mathematical Modeling after treatment.
2. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 1, with as follows
Feature：Comprise the following steps：
Step one, during real time phone call voip network parameter is acquired, obtains many set of network parameters；
Step 2, many set of network parameters to being collected in step one are pre-processed, and obtain corresponding multiple preprocessed data；
Step 3, the preprocessed data obtained to step 2 carry out dimension-reduction treatment using PCA principal component analytical methods, obtain low-dimensional
Data；
Step 4, to being calculated by the low-dimensional data obtained in step 3, obtain VoIP speech quality objective assessment scores；
So far, from step one to step 4, a kind of VoIP voice quality objective evaluation methods based on network parameter are completed.
3. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 1, with as follows
Feature：Step one is specially：Primary network parameter is obtained every t seconds (T >=t) in the T seconds interval in a communication process, is obtained
To T/t set of network parameters, packet loss, jitter value and delay value are included per set of network parameters.
4. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 1, with as follows
Feature：Step 2 is specially：Average, variance, most value, mode and the median for asking for the multi-group data under every kind of network parameter be
Main basic mathematical statistic, obtains N number of preprocessed data { X1,X2,…,Xn, n=1,2 ..., N, N are positive integer }.
5. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 1, with as follows
Feature：Step 3, specifically：Pretreated high dimensional data is converted to the low-dimensional data { Y for representing principal component1,Y2,…,
Ym, m=1,2 ..., M, M are positive integer, and M≤N }.
6. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 1, with as follows
Feature：Step 4, specially：The low-dimensional data that step 3 is exported is calculated using multiple linear regression model, is obtained T seconds
VoIP speech quality objective assessment score S, i.e. S=β in communication process0+β1Y1+β2Y2+…+βmYm, factor beta therein0,β1,
β2,…,βmObtained by precondition.
7. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 6, with as follows
Feature：Factor beta in step 40,β1,β2,…,βmObtained by precondition, specially：
Step 4.1 collection P (P >=200) bar time interval is the VoIP real time phone call speeches of T seconds, obtains a plurality of voice signal；And
It is acquired according to described in step one simultaneously, obtains the corresponding network parameter of every speech；
Every voice signal that step 4.2 is collected to step 4.1 again, using subjective evaluation method or method for objectively evaluating,
Obtain P speech quality fraction；Meanwhile, the P set of network parameters to collecting is processed according to step 2 and three, using step
Three low-dimensional datas for obtaining and speech quality MOS divide, and are trained by the method for multiple linear regression and obtained for VoIP speech matter
Measure the multiple linear regression model of objective evaluation, that is, the training coefficient in objective assessment score S computing formula；
Wherein, speech quality fraction, i.e. MOS point.
8. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 7, with as follows
Feature：P.800, the MOS subjective evaluation methods that subjective evaluation method can be advised using ITU-T；Method for objectively evaluating can be adopted
The POLQA method for objectively evaluating P.863 advised with ITU-T.
9. a kind of VoIP voice quality objective evaluation methods based on network parameter according to claim 7, with as follows
Feature：Wherein, the method for described multiple linear regression can use least square method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710126997.9A CN106803797A (en) | 2017-03-06 | 2017-03-06 | A kind of VoIP speech quality method for objectively evaluating based on network parameter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710126997.9A CN106803797A (en) | 2017-03-06 | 2017-03-06 | A kind of VoIP speech quality method for objectively evaluating based on network parameter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106803797A true CN106803797A (en) | 2017-06-06 |
Family
ID=58987891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710126997.9A Pending CN106803797A (en) | 2017-03-06 | 2017-03-06 | A kind of VoIP speech quality method for objectively evaluating based on network parameter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106803797A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108322346A (en) * | 2018-02-09 | 2018-07-24 | 山西大学 | A kind of voice quality assessment method based on machine learning |
| CN108933697A (en) * | 2018-06-30 | 2018-12-04 | 江苏有线数据网络有限责任公司 | A kind of broadband quality determining method |
| CN109120924A (en) * | 2018-10-30 | 2019-01-01 | 宁波菊风系统软件有限公司 | A kind of quality evaluating method of live video communication |
| CN110503981A (en) * | 2019-08-26 | 2019-11-26 | 苏州科达科技股份有限公司 | Without reference audio method for evaluating objective quality, device and storage medium |
| CN112822482A (en) * | 2020-12-31 | 2021-05-18 | 上海掌门科技有限公司 | Method and equipment for determining evaluation score of audio and video call |
| CN115963798A (en) * | 2023-02-23 | 2023-04-14 | 深圳市玄羽科技有限公司 | Equipment operation control method and system in industrial internet and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102044247A (en) * | 2009-10-10 | 2011-05-04 | 北京理工大学 | Objective evaluation method for VoIP speech |
| CN105282347A (en) * | 2014-07-22 | 2016-01-27 | 中国移动通信集团公司 | Method and device for evaluating voice quality |
-
2017
- 2017-03-06 CN CN201710126997.9A patent/CN106803797A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102044247A (en) * | 2009-10-10 | 2011-05-04 | 北京理工大学 | Objective evaluation method for VoIP speech |
| CN105282347A (en) * | 2014-07-22 | 2016-01-27 | 中国移动通信集团公司 | Method and device for evaluating voice quality |
Non-Patent Citations (3)
| Title |
|---|
| KAILASH CHANDRA MISHRA，PADMA CHARAN DAS: "Measuring Quality of Service of VoIP Based on Artificial Neural Network Approach", 《INTERNATIONAL JOURNAL OF ADVANCED RESEARCH IN COMPUTER SCIENCE AND SOFTWARE ENGINEERING》 * |
| WANG JING, LIU XIN, ZHAO XIAOHAN, LIU TONG, MENG LIUCHEN: "An Objective Measurement of Speech Quality in VoIP Based on Network Parameters", 《2016 IEEE 13TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS》 * |
| WEIWEI ZHANG, YONGYU CHANG, YITONG LIU,等: "Study on the QoE for VoIP Networks", 《JOURNAL OF NETWORKS》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108322346A (en) * | 2018-02-09 | 2018-07-24 | 山西大学 | A kind of voice quality assessment method based on machine learning |
| CN108322346B (en) * | 2018-02-09 | 2021-02-02 | 山西大学 | Voice quality evaluation method based on machine learning |
| CN108933697A (en) * | 2018-06-30 | 2018-12-04 | 江苏有线数据网络有限责任公司 | A kind of broadband quality determining method |
| CN108933697B (en) * | 2018-06-30 | 2021-02-19 | 江苏有线数据网络有限责任公司 | Broadband quality detection method |
| CN109120924A (en) * | 2018-10-30 | 2019-01-01 | 宁波菊风系统软件有限公司 | A kind of quality evaluating method of live video communication |
| CN109120924B (en) * | 2018-10-30 | 2020-06-02 | 宁波菊风系统软件有限公司 | Quality evaluation method for real-time video communication |
| CN110503981A (en) * | 2019-08-26 | 2019-11-26 | 苏州科达科技股份有限公司 | Without reference audio method for evaluating objective quality, device and storage medium |
| CN112822482A (en) * | 2020-12-31 | 2021-05-18 | 上海掌门科技有限公司 | Method and equipment for determining evaluation score of audio and video call |
| CN115963798A (en) * | 2023-02-23 | 2023-04-14 | 深圳市玄羽科技有限公司 | Equipment operation control method and system in industrial internet and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106803797A (en) | A kind of VoIP speech quality method for objectively evaluating based on network parameter | |
| CN104392390B (en) | A kind of secondary equipment of intelligent converting station appraisal procedure based on TOPSIS models | |
| CN111292020B (en) | Power grid real-time operation risk assessment method and system based on random forest | |
| CN105679335B (en) | Speech quality assessment method and system based on no line analysis | |
| CN109815855B (en) | Electronic equipment automatic test method and system based on machine learning | |
| CN105141446A (en) | Network equipment health degree assessment method determined based on objective weight | |
| CN115423009A (en) | Cloud edge coordination-oriented power equipment fault identification method and system | |
| CN111126489A (en) | Power transmission equipment state evaluation method based on ensemble learning | |
| KR101953558B1 (en) | Apparatus and Method for Fault Management of Smart Devices | |
| CN108959498A (en) | A kind of big data processing platform and its design method for health monitoring | |
| CN114255784A (en) | Substation equipment fault diagnosis method based on voiceprint recognition and related device | |
| CN111953531B (en) | Network fault analysis method and device | |
| CN117556359A (en) | Power distribution equipment health state evaluation system and method | |
| CN116975389A (en) | Transformer substation fault analysis method and device | |
| CN113033845B (en) | Construction method and device for power transmission resource co-construction and sharing | |
| CN103578043B (en) | Device, method and system for processing power grid data | |
| CN108322346A (en) | A kind of voice quality assessment method based on machine learning | |
| CN106376018A (en) | Multidimensional measurement data based frequency scrambling code PCI optimization apparatus and method thereof | |
| CN117596190B (en) | System and method for testing secondary signal transmission delay characteristics of transformer substation | |
| CN112465358A (en) | Voltage quality classification method and device based on support vector machine | |
| CN106341837A (en) | Inferior quality telephone traffic proportion value prediction method and device | |
| CN112884201B (en) | Non-intrusive load monitoring method, device and system based on artificial intelligence algorithm | |
| CN117354053B (en) | Network security protection method based on big data | |
| CN117148048B (en) | Power distribution network fault prediction method and system based on digital twin technology | |
| CN117793579B (en) | Metering equipment remote monitoring management method and system based on Internet of things |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170606 |
|
| WD01 | Invention patent application deemed withdrawn after publication |