CN109599121A - Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition - Google Patents
Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition Download PDFInfo
- Publication number
- CN109599121A CN109599121A CN201910007825.9A CN201910007825A CN109599121A CN 109599121 A CN109599121 A CN 109599121A CN 201910007825 A CN201910007825 A CN 201910007825A CN 109599121 A CN109599121 A CN 109599121A
- Authority
- CN
- China
- Prior art keywords
- sample
- drunk driving
- default
- standard
- test
- 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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification
- G10L17/26—Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification
- G10L17/02—Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
Abstract
The invention belongs to artificial intelligence fields, disclose a kind of drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition, this method comprises: obtaining test sample sound, default Eigenvalues analysis is carried out to the test sample sound, obtains test feature value;Standard Eigenvalue is obtained, the Standard Eigenvalue is obtained after carrying out default Eigenvalues analysis to N number of preset standard voice sample, wherein N is positive integer;The first confidence interval is constructed according to the Standard Eigenvalue;If the test feature value is not belonging to first confidence interval, determine the test sample sound for drunk driving sample.Drunk driving recognition methods provided by the invention can judge whether driver belongs to drunk driving by detecting the sound of speaking of driver, improve the efficiency of drunk driving identification.
Description
Technical field
The invention belongs to artificial intelligence fields, are to be related to a kind of drunk driving detection side based on Application on Voiceprint Recognition more specifically
Method, device, equipment and storage medium.
Background technique
It drives when intoxicated and refers to after drinking within 8 hours or drive vehicle after drunk within 24 hours.Statistics shows
After driver drives when intoxicated, a possibility that accident occurs is 15 times usually, 30% road traffic accident be by drivining a car under the influence of alcohol,
Caused by drunk driving.
Currently, being to be detected with air blowing type detector to the main detection method of drunk driving, by test drives, personnel blow
The content of alcohol judges whether the behavior of driver belongs to drunk driving behavior in implication out: when in vehicle operator's blood
Alcohol content be more than or equal to 20mg/100ml, be less than 80mg/100ml when, the driving behavior of driver, which belongs to, drinks
It drives;When the alcohol content in vehicle operator's blood is more than or equal to 80mg/100ml, the driving behavior of driver
Belong to drunk driving.However, the driver to drive when intoxicated, which often delays or mismatches, carries out air blowing detection, traffic is caused to be held
Method personnel detect an automobile drunk driving situation overlong time, be likely to result in other automobile storages drunk driving behavior can not
It takes into account, detection efficiency is lower.
Summary of the invention
The embodiment of the present invention provides a kind of drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition, with
Solve the problems, such as that the efficiency of current drunk driving detection is lower.
A kind of drunk driving detection method based on Application on Voiceprint Recognition, comprising:
Test sample sound is obtained, default Eigenvalues analysis is carried out to the test sample sound, obtains test feature value;
Standard Eigenvalue is obtained, the Standard Eigenvalue is that default characteristic value is carried out to N number of preset standard voice sample
It is obtained after analysis, wherein N is positive integer;
The first confidence interval is constructed according to the Standard Eigenvalue;
If the test feature value is not belonging to first confidence interval, determine the test sample sound for drunk driving sample
This.
A kind of drunk driving detection device based on Application on Voiceprint Recognition, comprising:
Test feature value obtains module, for obtaining test sample sound, carries out default spy to the test sample sound
Value indicative analysis, obtains test feature value;
Standard Eigenvalue obtains module, and for obtaining Standard Eigenvalue, the Standard Eigenvalue is to N number of preset standard
What sample sound obtain after default Eigenvalues analysis, wherein N is positive integer;
First confidence interval constructs module, for constructing the first confidence interval according to the Standard Eigenvalue;
Drunk driving sample determination module determines institute if being not belonging to first confidence interval for the test feature value
Stating test sample sound is drunk driving sample.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize the above-mentioned drunk driving based on Application on Voiceprint Recognition when executing the computer program
Detection method.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
Calculation machine program realizes the above-mentioned drunk driving detection method based on Application on Voiceprint Recognition when being executed by processor.
Above-mentioned drunk driving detection method, device, computer equipment and storage medium based on Application on Voiceprint Recognition is tested by obtaining
Sample sound carries out default Eigenvalues analysis to test sample sound, obtains test feature value;Then Standard Eigenvalue is obtained,
The first confidence interval is constructed according to Standard Eigenvalue, if test feature value is not belonging to the first confidence interval, discriminating test sound
Sample is drunk driving sample.By obtaining the sound of driver, the first confidence interval obtained with standard voice sample is compared
Compared with to judge whether driver belongs to drunk driving, driver delays or mismatch when can detect to avoid air blowing is carried out
The too long situation of caused detection time improves the efficiency of drunk driving identification.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is an application environment schematic diagram of the drunk driving detection method in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 2 is a flow chart of the drunk driving detection method in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 3 is another flow chart of the drunk driving detection method in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 4 is another flow chart of the drunk driving detection method in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 5 is another flow chart of the drunk driving detection method in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 6 is another flow chart of the drunk driving detection method in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 7 is a functional block diagram of the drunk driving detection device in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 8 is another functional block diagram of the drunk driving detection device in one embodiment of the invention based on Application on Voiceprint Recognition;
Fig. 9 is that test feature value obtains module in drunk driving detection device in one embodiment of the invention based on Application on Voiceprint Recognition
One functional block diagram;
Figure 10 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Drunk driving detection method provided by the present application based on Application on Voiceprint Recognition, can be applicable in the application environment such as Fig. 1,
In, client is communicated by network with server-side, and server-side obtains test sample sound by client, to test sound
Sample carries out default Eigenvalues analysis, obtains test feature value;Then server-side obtains Standard Eigenvalue, according to Standard Eigenvalue
The first confidence interval is constructed, if test feature value is not belonging to the first confidence interval, discriminating test sample sound is drunk driving sample,
And it will determine that result is exported to client.Wherein, client can be, but not limited to be various personal computers, laptop,
Smart phone, tablet computer and portable wearable device.Server-side can use independent server either multiple servers
The server cluster of composition is realized.
In one embodiment, it as shown in Fig. 2, providing a kind of drunk driving detection method based on Application on Voiceprint Recognition, answers in this way
It is illustrated, includes the following steps: for the server-side in Fig. 1
S10: obtaining test sample sound, carries out default Eigenvalues analysis to test sample sound, obtains test feature value.
Wherein, test sample sound refers to the sample sound of the driver of scene acquisition, can be obtained by client
After be sent to server-side.Optionally, client includes testing the acquisition module of sample sound, tests the acquisition module of sample sound
For example, recording device.When scene obtains, traffic police personnel can engage in the dialogue with driver, when driver speaks
When, the sound that traffic police personnel open client obtains module and obtains driver's one's voice in speech as test sound sample
This.Further, client can also use sound groove recognition technology in e, identify to the sound of typing, according to the result of identification
For the automatic sound for choosing driver as test sample sound, the sound for reducing ambient noise or other personnel detects drunk driving
Influence.
Wherein, default characteristic value refer to can reflect the corresponding behavior of sample sound whether be drunk driving characteristic value, can be with
Understand, after people drinks, stimulation of the throat of people by alcohol, it may occur that a degree of variation.And throat these variations can be with
Application on Voiceprint Recognition is carried out to test sample sound, is embodied by detecting default characteristic value.Optionally, default characteristic value can be frequency
At least one of in rate perturbation value, Shimmer value or standardization noise power value, or it is other to can reflect throat change
The acoustics index of change, the present embodiment are not specifically limited.
Specifically, server-side is used as test sample sound by the sound that client obtains driver, then server-side
Default Eigenvalues analysis is carried out to test sample sound by connection application on voiceprint recognition equipment, by the pre- of obtained test sample sound
If characteristic value is as test feature value.For example, the test sample sound obtained by client is sent to and is serviced by server-side
The application on voiceprint recognition equipment of end connection carries out default Eigenvalues analysis, to obtain test feature value.Optionally, application on voiceprint recognition equipment
It can be Electroglottography device (electroglottography), Electroglottography device can be to vocal cord vibration fundamental frequency, make rate, frequency
Rate perturbation value, Shimmer value, standardization noise power value and vocal cords outreach degree and larynx position height variation etc. features
Value is detected accordingly.In default Eigenvalues analysis, server-side will be tested in sample sound input Electroglottography device, can be obtained
It is worth accordingly to default characteristic value as test feature value.Alternatively it is also possible to application on voiceprint recognition equipment is integrated in client, visitor
Family end carries out default Eigenvalues analysis with application on voiceprint recognition equipment after getting test sample sound, obtains test feature value, then
The test feature value that analysis obtains is sent to server-side.
S20: obtaining Standard Eigenvalue, and Standard Eigenvalue is that default characteristic value is carried out to N number of preset standard voice sample
It is obtained after analysis, wherein N is positive integer.
Wherein, standard voice sample refers to the sample sound for having reached the personnel of drunk driving standard.It is alternatively possible to obtain logical
Cross blow detection or confirmation of the methods of taking a blood sample reach drunk driving standard personnel recording, regard these recording as standard voice
Sample.The personnel to drink in various degree can also be invited to be recorded to obtain standard voice sample, so as to drunk driving and
The standard voice sample of drunk driving makees further subdivision.Standard voice sample is N number of, it will be understood that the value of N is bigger, i.e. standard
Sample sound is more, and the accuracy of the drunk driving detection method based on Application on Voiceprint Recognition is higher.It is alternatively possible to collect in advance a large amount of
Standard voice sample is stored in the database of server-side, as preset standard voice sample.
Specifically, server-side is by obtaining the sample sound of N number of personnel for reaching drunk driving standard as standard voice sample,
Then default Eigenvalues analysis is carried out to this N number of standard voice sample, the characteristic value that analysis is obtained is as Standard Eigenvalue.
S30: the first confidence interval is constructed according to Standard Eigenvalue.
Wherein, confidence interval (Confidence interval, abbreviation CI) is to the overall parameter point for generating sample
Some unknown parameters ' value in cloth, the estimation provided with range format are estimated relative to point estimation with a sample statistic
Parameter value is counted, confidence interval has also contained the information of the accuracy of estimation.It is appreciated that can be improved by constructing confidence interval
Judge to test the accuracy whether sample sound belongs to drunk driving sample.
Specifically, server-side can obtain the variance of the Standard Eigenvalue of N number of preset standard voice sample, then root first
The first confidence interval is constructed according to obtained variance, so as to judge to test whether sample sound belongs to according to the first confidence interval
In drunk driving sample.
It is appreciated that server-side obtains Standard Eigenvalue after getting test sample sound again, according to Standard Eigenvalue
Construct the first confidence interval judge test sample sound whether belong to drunk driving sample, when standard voice sample changes,
Such as standard voice sample, when increasing, server-side can obtain a first new confidence according to the standard voice sample after variation
Section judges to test whether sample sound belongs to drunk driving sample further according to the first new confidence interval, convenient to be based on Application on Voiceprint Recognition
Drunk driving detection method update and upgrading.
In a specific embodiment, server-side can also first obtain the Standard Eigenvalue of N number of standard voice sample
Then average value sets default drunk driving threshold value according to obtained average value, if test feature value is more than or equal to default drunk driving
Threshold value, then discriminating test sample sound is drunk driving sample;If test feature value is less than default drunk driving threshold value, discriminating test sound
Sample is non-drunk driving sample.For example, if according to the average value of the lock in phenomenon value of Standard Eigenvalue by the default of lock in phenomenon value
Drunk driving threshold value is A, and test feature is worth corresponding lock in phenomenon value and is greater than A, then discriminating test sample sound is drunk driving sample.It is optional
Ground, default drunk driving threshold value can add and subtract an empirical value according to the average value of Standard Eigenvalue and obtain.
S40: if test feature value is not belonging to the first confidence interval, discriminating test sample sound is drunk driving sample.
Specifically, if test feature value is not belonging to the first confidence interval, server-side discriminating test sample sound is drunk driving
Sample, corresponding driver are drunk driving state, and the result of judgement is sent to client by server-side;If test feature value belongs to
First confidence interval, then server-side discriminating test sample sound is non-drunk driving sample, and corresponding driver is non-drunk driving state,
The result of judgement is sent to client by server-side.
It should be understood that since default characteristic value may include multiple detection contents, it can be according to multiple detection contents point
The first different confidence intervals is not constructed.Optionally, server-side can be set when each test feature value belongs to each detection
When corresponding first confidence interval of content, discriminating test sample sound is drunk driving sample;The survey when predetermined fraction can also be set
When examination characteristic value belongs to corresponding first confidence interval, discriminating test sample sound is drunk driving sample, other parts test feature
Value is then used as reference data.
In the corresponding embodiment of Fig. 2, sample sound is tested by obtaining, default characteristic value is carried out to test sample sound
Analysis, obtains test feature value;Then Standard Eigenvalue is obtained, the first confidence interval is constructed according to Standard Eigenvalue, if test
Characteristic value is not belonging to the first confidence interval, then discriminating test sample sound is drunk driving sample.By obtaining the sound of driver,
It is compared with the first confidence interval that standard voice sample obtains, to judge whether driver belongs to drunk driving, can keep away
Exempt from blow driver when detecting and delay or mismatch the too long situation of caused detection time, improves drunk driving identification
Efficiency.
In one embodiment, as shown in figure 3, default characteristic value is lock in phenomenon value, Shimmer value and standardization noise
At least one of energy value, it is before step S20, i.e., provided in this embodiment before the step of obtaining Standard Eigenvalue
Drunk driving detection method based on Application on Voiceprint Recognition is further comprising the steps of:
S51: standard voice sample is obtained.
Wherein, the method for server-side acquisition standard voice sample is identical as the method for step S20, and which is not described herein again.
S52: Application on Voiceprint Recognition is carried out to each standard voice sample, obtains the default characteristic value of each standard voice sample.
Wherein, Application on Voiceprint Recognition refers to that standard voice sample, which is sent to application on voiceprint recognition equipment, carries out default Eigenvalues analysis,
Default characteristic value is at least one of lock in phenomenon value, Shimmer value and standardization noise power value.Optionally, vocal print is known
Other equipment is Electroglottography device, is also possible to other application on voiceprint recognition equipment, is not specifically limited here.
Lock in phenomenon (jitter) value is to describe the physical quantity of basic frequency variation between sound wave adjacent periods, main to reflect
Coarse sound path degree also reflects hoarse sound path degree.The functional status of jitter value and vocal area in voice signal is consistent, i.e.,
Under normal circumstances, the identical sound wave of frequency during voice week is more, and the sound wave of different frequency is seldom, and lock in phenomenon value is very at this time
It is small;And after people drinks, the functional status of stimulation of the throat of people by alcohol, vocal area can change, and keep sound coarse,
Jitter value increases.
Shimmer (shimmer) value describes the variation of wave amplitude between adjacent periods, mainly reflects hoarse sound path degree.
After people drinks, stimulation of the throat of people by alcohol, the functional status of vocal area can change, and make hoarseness,
Shimmer value increases.It is appreciated that Jitter value and shimmer value all reflect the stability of vocal cord vibration, the bigger explanation of value
The minor change that acoustic signal occurs in voiced process is bigger.
Standardization noise energy (NNE) value be calculate sounding when due to glottis it is non fully-closed caused by glottis noise energy
Amount, it is main to reflect breathiness degree, also reflect the closing degree of hoarse sound path degree and glottis, after people drinks, the throat of people by
Functional status to the stimulation of alcohol, vocal area can change, and increase the breathiness in sound, so that NNE value increases.
Specifically, server-side carries out Application on Voiceprint Recognition to each standard voice sample in N number of standard voice sample, right
At least one in jitter value, shimmer value or NNE value is preset characteristic value and is analyzed, and each standard voice sample is obtained
At least one in jitter value, shimmer value or NNE value presets characteristic value.
S53: using the default characteristic value of each standard voice sample as the Standard Eigenvalue of corresponding standard voice sample.
Specifically, every jitter value, shimmer value or the NNE value for each standard voice sample that server-side will acquire
In at least one default characteristic value be stored in the database of server-side, the standard feature as corresponding standard voice sample
Value.It is appreciated that can more embody drunk driving when Standard Eigenvalue includes three values in jitter value, shimmer value and NNE value
The accuracy rate of the sound characteristic of personnel, the drunk driving detection method based on Application on Voiceprint Recognition is higher.
In the corresponding embodiment of Fig. 3, by obtaining standard voice sample, then to each standard voice sample carry out sound
Line identification, obtains the default characteristic value of each standard voice sample, finally makees the default characteristic value of each standard voice sample
For the Standard Eigenvalue of corresponding standard voice sample.By lock in phenomenon value to standard voice sample, Shimmer value or
The default characteristic value of at least one in standardization noise power value is analyzed, to judge to test whether sample sound belongs to drunk driving sample
Originally the foundation that data are supported and judged is provided.In addition, due to lock in phenomenon value, Shimmer value or standardization noise power value
It can be well reflected the sound characteristic of drunk driver, therefore can be improved and judge whether driver is the accurate of drunk driver
Degree.
In one embodiment, as shown in figure 4, in step slo, that is, test sample sound is obtained, to test sample sound
Default Eigenvalues analysis is carried out, test feature value is obtained, can specifically include following steps:
S11: it is surveyed based on being obtained under the conditions of at least one in default pronunciation phonemes, default phonation time or default environmental condition
Try sample sound.
It is appreciated that better Application on Voiceprint Recognition effect in order to obtain, obtain test sample sound when, when can be to sounding
Between, at least one condition is provided accordingly in pronunciation phonemes or environmental condition.Optionally, default pronunciation phonemes can be member
Sound;Vowel, e.g. a, u, o etc., also known as vowel are one kind of phoneme, opposite with consonant;Vowel is during the pronunciation process by gas
For stream by the sound of the unobstructed sending in oral cavity, different vowels is as caused by the different shape in oral cavity, it will be understood that is passed through
Vowel can preferably reflect the feature of sound.Optionally, default phonation time can be 3~5 seconds, keep sound complete
Typing, the default characteristic value of sound can completely be embodied.Optionally, default environmental condition can be ambient noise control
System is below 45dB SPL.Since on-the-spot test would generally be influenced by live noise, the realization for presetting environmental condition can be with
It is that driver is brought to environment appropriate (such as car or traffic police office of traffic police), it can also be in default pronunciation phonemes
And/or default phonation time is determined as on the basis of drunk driving, then makees further confirmation under default environmental condition.
Specifically, server-side at least one condition in default pronunciation phonemes, default phonation time and default environmental condition
Sample sound is tested in lower acquisition.It is appreciated that the condition met is more, the accuracy of drunk driving identification is higher.
S12: Application on Voiceprint Recognition is carried out to test sample sound, obtains the default characteristic value of test sample sound.
Wherein, Application on Voiceprint Recognition refers to that test sample sound, which is sent to application on voiceprint recognition equipment, carries out default Eigenvalues analysis,
Default characteristic value is at least one of lock in phenomenon value, Shimmer value and standardization noise power value.Optionally, vocal print is known
Other equipment is Electroglottography device, is also possible to other application on voiceprint recognition equipment, is not specifically limited here.
Specifically, server-side carries out in jitter value, shimmer value or NNE value at least the test sample sound of acquisition
It is pre- to obtain at least one in the jitter value, shimmer value or NNE value of test sample sound for the analysis of one default characteristic value
If characteristic value.
S13: the default characteristic value of sample sound will be tested as test feature value.
Specifically, server-side will be at least one in the jitter value, shimmer value or NNE value of obtained test sample sound
A default characteristic value is stored in the database of server-side as test feature value.Optionally, Standard Eigenvalue and test are special
Jitter value, shimmer value and the NNE value that value indicative includes are corresponding.For example, if Standard Eigenvalue include jitter value and
Shimmer value, then test feature value also includes jitter value and shimmer value.
In the corresponding embodiment of Fig. 4, by based in default pronunciation phonemes, default phonation time or default environmental condition
Test sample sound is obtained under the conditions of at least one, Application on Voiceprint Recognition then is carried out to test sample sound, obtains test sound sample
This default characteristic value;The default characteristic value of sample sound will finally be tested as test feature value.By to test sound sample
This when, carries out pronunciation phonemes, phonation time or environmental condition and is standardized, and can make the test sample sound obtained that can more reflect
Default characteristic value is embodied, judges to test the accuracy whether sample sound belongs to drunk driving sample to improve.
In one embodiment, as shown in figure 5, in step s 30, i.e., constructing the first confidence interval according to Standard Eigenvalue,
It can specifically include following steps:
S31: the variance of N number of Standard Eigenvalue is obtained.
Specifically, if N number of Standard Eigenvalue is expressed as (x1,x2,x3…xN), server-side can be calculated N number of with following formula
The variance of the corresponding Standard Eigenvalue of standard voice sample:
Wherein, x refers to that Standard Eigenvalue, μ are (x1,x2,x3…xN) average value.
S32: the first confidence interval is constructed according to the variance of N number of Standard Eigenvalue.
Specifically, server-side can choose a significance according to the input of client, further according to the significant of selection
Property level construct the first confidence interval.Wherein, significance is that estimation population parameter is fallen in a certain section, may be made mistakes
Probability, i.e., the probability that may make mistakes when constructing confidence interval by the significance.For example, if the first confidence of building
Section is B, if test feature value is not fallen in this first confidence interval of B, server-side is determined as sample sound is tested
Non- drunk driving sample, but actually test sample sound is drunk driving sample, then to make mistakes.Optionally, significance can take
0.05,0.01, by taking significance takes 0.05 as an example, correspondingly, fiducial range 95%, then the first confidence interval is constructed are as follows:I.e. withThe first confidence interval is constructed for error range.
Specifically, the test feature value for testing sample sound is compared by server-side with the first confidence interval of building,
If test feature value is fallen into the first confidence interval of building, discriminating test sample sound is non-drunk driving sample, and by non-wine
The judgement result driven is sent to client;If test feature value is fallen in outside the first confidence interval of building, discriminating test sound
Sample is drunk driving sample, and the judgement result of drunk driving is sent to client.
In the corresponding embodiment of Fig. 5, by obtaining the variance of N number of Standard Eigenvalue, further according to N number of Standard Eigenvalue
Variance constructs the first confidence interval, can provide the foundation of judgement for judgement test sample sound for drunk driving sample.In addition, passing through
Confidence interval is constructed to realize the identification of drunk driving sample, the accuracy of drunk driving identification can be improved.
In one embodiment, as shown in fig. 6, after the step s 40, if being not belonging to the first confidence area in test feature value
Between, then discriminating test sample sound be drunk driving sample the step of after, it is provided in this embodiment based on Application on Voiceprint Recognition drunk driving inspection
Survey method is further comprising the steps of:
S61: if the data variance of test feature value and N number of Standard Eigenvalue is less than default significance, by phase
The drunk driving Sample preservation answered is into database as new standard voice sample.
Wherein, the data variance of test feature value and Standard Eigenvalue can be measured with significance, this implementation
In example, default significance can be set according to actual needs, can take 0.05 or 0.01.It is appreciated that when default
When significance difference, the confidence interval of building is also different.Determine due to being necessary to ensure that server-side will test sample sound
It is lower for the probability made mistakes when drunk driving sample, default significance can be taken as to 0.01, i.e. server-side is by drunk driving sample
Default significance is taken as 0.01 when saving in the database as new standard voice sample, and judges to test in server-side
Sample sound is that the significance of drunk driving sample is taken as 0.05.
Specifically, it if the data variance of test feature value and N number of Standard Eigenvalue is less than default significance, takes
Corresponding drunk driving sample (test sample sound) can be stored in the database of server-side as new standard voice sample by business end
This.It is appreciated that the quantity of the standard voice sample in database is more, the drunk driving detection method based on Application on Voiceprint Recognition is more quasi-
Really.
S62: new standard voice sample is incorporated in N number of standard voice sample, obtains updated standard voice sample.
Specifically, new standard voice sample is incorporated in original N number of standard voice sample by server-side, is obtained updated
Standard voice sample.It is appreciated that new standard voice sample can be it is multiple, i.e., synchronization has n drunk driving sample to switch to
New standard voice sample.If such as new standard voice number of samples is n, has N+ in the database of server-side after increasing
N standard voice sample.Optionally, server-side can also be updated standard voice sample according to prefixed time interval, in advance
If time interval is, for example, one day, one week or one month etc., specifically without limitation.
S63: obtaining the variance of updated standard voice sample, is constructed according to the variance of updated standard voice sample
Second confidence interval.
Specifically, server-side can obtain the variance of updated standard voice sample, then root according to the method for step S31
Confidence interval is rebuild according to the variance of updated standard voice sample, obtains the second confidence interval.When needs judge newly
When whether test sample sound belongs to drunk driving sample, is judged with the second confidence interval newly constructed, judge new test sound
Whether the corresponding test feature value of sound sample belongs to the second confidence interval, if the test feature value is not belonging to the second confidence interval,
Then server-side determines new test sample sound for drunk driving sample.It is appreciated that passing through the mark for constantly expanding server database
Quasi- sample sound can make the drunk driving detection method based on Application on Voiceprint Recognition more accurate.
In the corresponding embodiment of Fig. 6, if the data variance of test feature value and N number of Standard Eigenvalue is less than default show
Work property is horizontal, then by corresponding drunk driving Sample preservation into database as new standard voice sample;Then by new standard
Sample sound is incorporated in N number of standard voice sample, obtains updated standard voice sample, then obtain updated standard voice
The variance of sample constructs the second confidence interval according to the variance of updated standard voice sample.It is default significant by that will be less than
Property horizontal drunk driving sample be incorporated in standard voice sample, the quantity of standard voice sample can be expanded;And according to expansion after
The confidence interval that standard voice sample obtains goes to judge whether test sample sound belongs to drunk driving sample, and drunk driving can be continuously improved
The accuracy of identification.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of drunk driving detection device based on Application on Voiceprint Recognition is provided, it should the drunk driving based on Application on Voiceprint Recognition
Drunk driving detection method in detection device and above-described embodiment based on Application on Voiceprint Recognition corresponds.As shown in fig. 7, vocal print should be based on
The drunk driving detection device of identification includes that test feature value obtains module 10, Standard Eigenvalue obtains module 20, the first confidence interval
Construct module 30 and drunk driving sample determination module 40.Detailed description are as follows for each functional module:
Test feature value obtains module 10, for obtaining test sample sound, presets to the test sample sound
Eigenvalues analysis obtains test feature value;
Standard Eigenvalue obtains module 20, and for obtaining Standard Eigenvalue, the Standard Eigenvalue is to N number of preset mark
What quasi- sample sound obtain after default Eigenvalues analysis, wherein N is positive integer;
First confidence interval constructs module 30, for constructing the first confidence interval according to the Standard Eigenvalue;
Drunk driving sample determination module 40 determines if being not belonging to first confidence interval for the test feature value
The test sample sound is drunk driving sample.
Further, preset characteristic value be lock in phenomenon value, Shimmer value and standardization noise power value at least
One, as shown in figure 8, the drunk driving detection device provided in this embodiment based on Application on Voiceprint Recognition further includes default Eigenvalues analysis mould
Block 50, presetting Eigenvalues analysis module 50 includes master sample acquiring unit 51, the first Application on Voiceprint Recognition unit 52, Standard Eigenvalue
Determination unit 53.
Master sample acquiring unit 51, for obtaining standard voice sample;
First Application on Voiceprint Recognition unit 52 obtains each standard sound for carrying out Application on Voiceprint Recognition to each standard voice sample
The default characteristic value of sound sample;
Standard Eigenvalue determination unit 53, for using the default characteristic value of each standard voice sample as corresponding standard
The Standard Eigenvalue of sample sound.
Further, as shown in figure 9, it further includes test sample acquiring unit 11, second that test feature value, which obtains module 10,
Application on Voiceprint Recognition unit 12 and test feature value determination unit 13.
Test sample acquiring unit 11, for based in default pronunciation phonemes, default phonation time or default environmental condition
Test sample sound is obtained under the conditions of at least one;
Second vocal print recognition unit 12 obtains test sample sound for carrying out Application on Voiceprint Recognition to test sample sound
Default characteristic value;
Test feature value determination unit 13, for the default characteristic value of sample sound will to be tested as test feature value.
Further, the first confidence interval building module 30 is also used to:
Obtain the variance of N number of Standard Eigenvalue;
The first confidence interval is constructed according to the variance of N number of Standard Eigenvalue.
Further, the drunk driving detection device provided in this embodiment based on Application on Voiceprint Recognition further includes the second confidence interval structure
Model block, wherein the second confidence interval building module is specifically used for:
If the data variance of test feature value and N number of Standard Eigenvalue is less than default significance, will be corresponding
Drunk driving Sample preservation is into database as new standard voice sample;
New standard voice sample is incorporated in N number of standard voice sample, updated standard voice sample is obtained;
The variance for obtaining updated standard voice sample, according to the variance of updated standard voice sample building second
Confidence interval.
Specific restriction about the drunk driving detection device based on Application on Voiceprint Recognition may refer to know above for based on vocal print
The restriction of other drunk driving detection method, details are not described herein.Each mould in the above-mentioned drunk driving detection device based on Application on Voiceprint Recognition
Block can be realized fully or partially through software, hardware and combinations thereof.Above-mentioned each module can be embedded in the form of hardware or independence
In processor in computer equipment, it can also be stored in a software form in the memory in computer equipment, in order to
Processor, which calls, executes the corresponding operation of the above modules.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 10.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing test sample sound, standard voice sample, test feature value, Standard Eigenvalue and drunk driving sample
This etc..The network interface of the computer equipment is used to communicate with external terminal by network connection.The computer program is located
It manages when device executes to realize a kind of drunk driving detection method based on Application on Voiceprint Recognition.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor perform the steps of when executing computer program
Test sample sound is obtained, default Eigenvalues analysis is carried out to test sample sound, obtains test feature value;
Standard Eigenvalue is obtained, Standard Eigenvalue is that default Eigenvalues analysis is carried out to N number of preset standard voice sample
It obtains afterwards, wherein N is positive integer;
The first confidence interval is constructed according to Standard Eigenvalue;
If test feature value is not belonging to the first confidence interval, discriminating test sample sound is drunk driving sample.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Test sample sound is obtained, default Eigenvalues analysis is carried out to test sample sound, obtains test feature value;
Standard Eigenvalue is obtained, Standard Eigenvalue is that default Eigenvalues analysis is carried out to N number of preset standard voice sample
It obtains afterwards, wherein N is positive integer;
The first confidence interval is constructed according to Standard Eigenvalue;
If test feature value is not belonging to the first confidence interval, discriminating test sample sound is drunk driving sample.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of drunk driving detection method based on Application on Voiceprint Recognition characterized by comprising
Test sample sound is obtained, default Eigenvalues analysis is carried out to the test sample sound, obtains test feature value;
Standard Eigenvalue is obtained, the Standard Eigenvalue is that default Eigenvalues analysis is carried out to N number of preset standard voice sample
It obtains afterwards, wherein N is positive integer;
The first confidence interval is constructed according to the Standard Eigenvalue;
If the test feature value is not belonging to first confidence interval, determine the test sample sound for drunk driving sample.
2. the drunk driving detection method based on Application on Voiceprint Recognition as described in claim 1, which is characterized in that the default characteristic value is
At least one of lock in phenomenon value, Shimmer value and standardization noise power value;
Before the acquisition Standard Eigenvalue the step of, the drunk driving detection method based on Application on Voiceprint Recognition further include:
Obtain the standard voice sample;
Application on Voiceprint Recognition is carried out to each standard voice sample, obtains the default feature of each standard voice sample
Value;
It is special using the default characteristic value of each standard voice sample as the standard of the corresponding standard voice sample
Value indicative.
3. the drunk driving detection method based on Application on Voiceprint Recognition as claimed in claim 2, which is characterized in that sound is tested in the acquisition
Sample carries out default Eigenvalues analysis to the test sample sound, obtains test feature value, comprising:
Sound sample is tested based on obtaining under the conditions of at least one in default pronunciation phonemes, default phonation time or default environmental condition
This;
Application on Voiceprint Recognition is carried out to the test sample sound, obtains the default characteristic value of the test sample sound;
Using the default characteristic value of the test sample sound as test feature value.
4. the drunk driving detection method based on Application on Voiceprint Recognition as claimed in claim 3, which is characterized in that described according to the standard
Characteristic value constructs the first confidence interval, comprising:
Obtain the variance of N number of Standard Eigenvalue;
The first confidence interval is constructed according to the variance of N number of Standard Eigenvalue.
5. the drunk driving detection method based on Application on Voiceprint Recognition as claimed in claim 4, which is characterized in that if in the test
It is described after the step of characteristic value is not belonging to first confidence interval, then determines the test sample sound for drunk driving sample
Drunk driving detection method based on Application on Voiceprint Recognition further include:
If the data variance of the test feature value and N number of Standard Eigenvalue is less than default significance, by phase
The drunk driving Sample preservation answered is into database as new standard voice sample;
The new standard voice sample is incorporated in N number of standard voice sample, updated standard voice sample is obtained;
The variance for obtaining updated standard voice sample, according to the variance of updated standard voice sample building second
Confidence interval.
6. a kind of drunk driving detection device based on Application on Voiceprint Recognition characterized by comprising
Test feature value obtains module, for obtaining test sample sound, carries out default characteristic value to the test sample sound
Analysis, obtains test feature value;
Standard Eigenvalue obtains module, and for obtaining Standard Eigenvalue, the Standard Eigenvalue is to N number of preset standard voice
What sample obtain after default Eigenvalues analysis, wherein N is positive integer;
First confidence interval constructs module, for constructing the first confidence interval according to the Standard Eigenvalue;
Drunk driving sample determination module determines the survey if being not belonging to first confidence interval for the test feature value
Examination sample sound is drunk driving sample.
7. the drunk driving detection device based on Application on Voiceprint Recognition as claimed in claim 6, which is characterized in that the default characteristic value is
At least one of lock in phenomenon value, Shimmer value and standardization noise power value;
The drunk driving detection device based on Application on Voiceprint Recognition further includes default Eigenvalues analysis module, the default Eigenvalues analysis
Module includes master sample acquiring unit, the first Application on Voiceprint Recognition unit, Standard Eigenvalue determination unit;
The master sample acquiring unit, for obtaining the standard voice sample;
The first Application on Voiceprint Recognition unit obtains each described for carrying out Application on Voiceprint Recognition to each standard voice sample
The default characteristic value of standard voice sample;
The Standard Eigenvalue determination unit, for using the default characteristic value of each standard voice sample as correspondence
The standard voice sample Standard Eigenvalue.
8. the drunk driving detection device based on Application on Voiceprint Recognition as claimed in claim 6, which is characterized in that the test feature value obtains
Modulus block includes test sample acquiring unit, the second vocal print recognition unit and test feature value determination unit;
The test sample acquiring unit, for being based in default pronunciation phonemes, default phonation time or default environmental condition extremely
Test sample sound is obtained under the conditions of one few;
The second vocal print recognition unit obtains the test sound for carrying out Application on Voiceprint Recognition to the test sample sound
The default characteristic value of sample;
The test feature value determination unit, for using the default characteristic value of the test sample sound as test feature
Value.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
Drunk driving detection method described in 5 any one based on Application on Voiceprint Recognition.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realization is as described in any one of claim 1 to 5 based on the drunk driving of Application on Voiceprint Recognition when the computer program is executed by processor
Detection method.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910007825.9A CN109599121A (en) | 2019-01-04 | 2019-01-04 | Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition |
PCT/CN2019/089161 WO2020140376A1 (en) | 2019-01-04 | 2019-05-30 | Drunk driving detection method and apparatus based on voiceprint recognition, and device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910007825.9A CN109599121A (en) | 2019-01-04 | 2019-01-04 | Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109599121A true CN109599121A (en) | 2019-04-09 |
Family
ID=65964943
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910007825.9A Pending CN109599121A (en) | 2019-01-04 | 2019-01-04 | Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109599121A (en) |
WO (1) | WO2020140376A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020140376A1 (en) * | 2019-01-04 | 2020-07-09 | 平安科技(深圳)有限公司 | Drunk driving detection method and apparatus based on voiceprint recognition, and device and storage medium |
CN113590868A (en) * | 2021-06-11 | 2021-11-02 | 深圳供电局有限公司 | Sound threshold updating method and device, computer equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070005357A1 (en) * | 2005-06-29 | 2007-01-04 | Rosalyn Moran | Telephone pathology assessment |
JP2009096327A (en) * | 2007-10-17 | 2009-05-07 | Xanavi Informatics Corp | Drunken state detection device |
US20100204993A1 (en) * | 2006-12-19 | 2010-08-12 | Robert Vogt | Confidence levels for speaker recognition |
CN101887722A (en) * | 2009-06-18 | 2010-11-17 | 博石金(北京)信息技术有限公司 | Rapid voiceprint authentication method |
US20140379332A1 (en) * | 2011-06-20 | 2014-12-25 | Agnitio, S.L. | Identification of a local speaker |
CN104505102A (en) * | 2014-12-31 | 2015-04-08 | 宇龙计算机通信科技(深圳)有限公司 | Method and device for examining physical conditions |
CN108269574A (en) * | 2017-12-29 | 2018-07-10 | 安徽科大讯飞医疗信息技术有限公司 | Voice signal processing method and device, storage medium and electronic equipment |
CN109102825A (en) * | 2018-07-27 | 2018-12-28 | 科大讯飞股份有限公司 | One kind is drunk condition detection method and device |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102194134B (en) * | 2010-03-01 | 2013-06-05 | 中国科学院自动化研究所 | Biological feature recognition performance index prediction method based on statistical learning |
CN103971680B (en) * | 2013-01-24 | 2018-06-05 | 华为终端(东莞)有限公司 | A kind of method, apparatus of speech recognition |
CN107424614A (en) * | 2017-07-17 | 2017-12-01 | 广东讯飞启明科技发展有限公司 | A kind of sound-groove model update method |
CN109599121A (en) * | 2019-01-04 | 2019-04-09 | 平安科技(深圳)有限公司 | Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition |
-
2019
- 2019-01-04 CN CN201910007825.9A patent/CN109599121A/en active Pending
- 2019-05-30 WO PCT/CN2019/089161 patent/WO2020140376A1/en active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070005357A1 (en) * | 2005-06-29 | 2007-01-04 | Rosalyn Moran | Telephone pathology assessment |
US20100204993A1 (en) * | 2006-12-19 | 2010-08-12 | Robert Vogt | Confidence levels for speaker recognition |
JP2009096327A (en) * | 2007-10-17 | 2009-05-07 | Xanavi Informatics Corp | Drunken state detection device |
CN101887722A (en) * | 2009-06-18 | 2010-11-17 | 博石金(北京)信息技术有限公司 | Rapid voiceprint authentication method |
US20140379332A1 (en) * | 2011-06-20 | 2014-12-25 | Agnitio, S.L. | Identification of a local speaker |
CN104505102A (en) * | 2014-12-31 | 2015-04-08 | 宇龙计算机通信科技(深圳)有限公司 | Method and device for examining physical conditions |
CN108269574A (en) * | 2017-12-29 | 2018-07-10 | 安徽科大讯飞医疗信息技术有限公司 | Voice signal processing method and device, storage medium and electronic equipment |
CN109102825A (en) * | 2018-07-27 | 2018-12-28 | 科大讯飞股份有限公司 | One kind is drunk condition detection method and device |
Non-Patent Citations (1)
Title |
---|
冯彦等: "吸烟饮酒对嗓音影响的客观定量研究", 中国药物与临床, vol. 7, no. 10, pages 749 - 750 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020140376A1 (en) * | 2019-01-04 | 2020-07-09 | 平安科技(深圳)有限公司 | Drunk driving detection method and apparatus based on voiceprint recognition, and device and storage medium |
CN113590868A (en) * | 2021-06-11 | 2021-11-02 | 深圳供电局有限公司 | Sound threshold updating method and device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
WO2020140376A1 (en) | 2020-07-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11538472B2 (en) | Processing speech signals in voice-based profiling | |
US7647224B2 (en) | Apparatus, method, and computer program product for speech recognition | |
US8612234B2 (en) | Multi-state barge-in models for spoken dialog systems | |
US8000971B2 (en) | Discriminative training of multi-state barge-in models for speech processing | |
US20080046241A1 (en) | Method and system for detecting speaker change in a voice transaction | |
US8447603B2 (en) | Rating speech naturalness of speech utterances based on a plurality of human testers | |
US7359856B2 (en) | Speech detection system in an audio signal in noisy surrounding | |
KR20190068830A (en) | Apparatus and method for determining recommendation reliability based on environment of vehicle | |
AU2009295251B2 (en) | Method of analysing an audio signal | |
CN109599121A (en) | Drunk driving detection method, device, equipment and storage medium based on Application on Voiceprint Recognition | |
Baumeister et al. | The influence of alcoholic intoxication on the fundamental frequency of female and male speakers | |
CN110459242A (en) | Change of voice detection method, terminal and computer readable storage medium | |
CN110047474A (en) | A kind of English phonetic pronunciation intelligent training system and training method | |
CN112885336A (en) | Training and recognition method and device of voice recognition system, and electronic equipment | |
Saxon et al. | Objective measures of plosive nasalization in hypernasal speech | |
Yap et al. | Glottal features for speech-based cognitive load classification | |
Wokurek et al. | Automated corpus based spectral measurement of voice quality parameters | |
KR101892736B1 (en) | Apparatus and method for utterance verification based on word duration | |
KR100298177B1 (en) | Method for construction anti-phone model and method for utterance verification based on anti-phone medel | |
US11024302B2 (en) | Quality feedback on user-recorded keywords for automatic speech recognition systems | |
Mills | Cues to voicing contrasts in whispered Scottish obstruents | |
CN111341298A (en) | Speech recognition algorithm scoring method | |
Wang et al. | Improved Mandarin speech recognition by lattice rescoring with enhanced tone models | |
Liu | Word fragments identification using acoustic-prosodic features in conversational speech | |
Yin et al. | Investigating speech features and automatic measurement of cognitive load |
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 |