CN109036436A - A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system - Google Patents
A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system Download PDFInfo
- Publication number
- CN109036436A CN109036436A CN201811089314.8A CN201811089314A CN109036436A CN 109036436 A CN109036436 A CN 109036436A CN 201811089314 A CN201811089314 A CN 201811089314A CN 109036436 A CN109036436 A CN 109036436A
- Authority
- CN
- China
- Prior art keywords
- voiceprint
- information
- voice print
- label information
- vocal print
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 76
- 230000001755 vocal effect Effects 0.000 claims abstract description 81
- 238000003909 pattern recognition Methods 0.000 claims abstract description 28
- 239000000284 extract Substances 0.000 claims abstract description 18
- 238000004590 computer program Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011840 criminal investigation Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000000116 mitigating effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- 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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/04—Training, enrolment or model building
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system, including obtain voiceprint, and extract the vocal print feature of voiceprint；Pattern match is carried out to vocal print feature using the pattern recognition model pre-established, obtains label information corresponding with vocal print feature；It stores label information is corresponding with voiceprint into voice print database, to be screened according to label information to voiceprint in Application on Voiceprint Recognition.It not only include voiceprint in the voice print database established by the voice print database method for building up in the application, it further include label information corresponding with voiceprint, to which the range of search of voiceprint can be reduced according to corresponding label information in subsequent Application on Voiceprint Recognition, to improve the recognition efficiency of Application on Voiceprint Recognition, Application on Voiceprint Recognition the time it takes is reduced.
Description
Technical field
The present embodiments relate to sound groove recognition technology in e fields, more particularly to a kind of voice print database method for building up, sound
Line recognition methods, apparatus and system.
Background technique
Vocal print is the sound wave spectrum for the carrying verbal information that electricity consumption acoustic instrument is shown.Modern scientific research shows vocal print
Not only there is specificity, but also have the characteristics of relative stability, Application on Voiceprint Recognition is one according to reflecting speaker in speech waveform
The speech parameter of physiology and behavioural characteristic, the technology of automatic identification speaker's identity, and for very extensive.
When carrying out Application on Voiceprint Recognition, it is normally based on the voice print database pre-established and is identified, it is in the prior art
Voice print database is only stored with voiceprint corresponding with collected personnel when establishing, and is passing through the voice print database carry out sound
When line identifies, need one by one to be matched all voiceprints stored in vocal print to be identified and voice print database, due to sound
The voiceprint data volume stored in line database is larger, therefore spends Application on Voiceprint Recognition using database in the prior art
Time it is longer, reduce Application on Voiceprint Recognition efficiency.
In consideration of it, how to provide it is a kind of solve the voice print database method for building up of above-mentioned technical problem, method for recognizing sound-groove,
Apparatus and system becomes those skilled in the art's problem to be solved.
Summary of the invention
The purpose of the embodiment of the present invention is that providing a kind of voice print database method for building up, method for recognizing sound-groove, device and being
System is imitated by the identification that the voice print database established in the application is conducive to improve when carrying out Application on Voiceprint Recognition Application on Voiceprint Recognition
Rate reduces Application on Voiceprint Recognition the time it takes.
In order to solve the above technical problems, the embodiment of the invention provides a kind of voice print database method for building up, comprising:
Voiceprint is obtained, and extracts the vocal print feature of the voiceprint；
Pattern match is carried out to the vocal print feature using the pattern recognition model pre-established, is obtained special with the vocal print
Levy corresponding label information；
Store the label information is corresponding with the voiceprint into voice print database, so as in Application on Voiceprint Recognition according to
Voiceprint is screened according to the label information.
Optionally, described that pattern match is carried out to the vocal print feature using the pattern recognition model pre-established, it obtains
The process of label information corresponding with the vocal print feature includes:
The vocal print feature is analyzed using the pattern recognition model pre-established, is obtained and the vocal print feature pair
The pattern information answered；The pattern information includes one of language information, age information, gender information or native place information or more
The combination of kind；
Label information corresponding with the vocal print feature is generated according to the pattern information.
Optionally, the method for building up of the pattern recognition model includes:
Analyzing and training is carried out using preset algorithm pair vocal print feature corresponding with multiple history voiceprints, obtains institute
State pattern recognition model；The preset algorithm includes that languages recognizer, gender recognizer, age recognizer or native place are known
One of other algorithm or a variety of combinations.
Optionally, the vocal print feature includes one of the sound intensity, loudness, pitch, pitch period or fundamental frequency or more
Kind.
The embodiment of the present invention provides a kind of voice print database accordingly and establishes device, comprising:
First extraction module for obtaining voiceprint, and extracts the vocal print feature of the voiceprint；
Processing module is obtained for carrying out pattern match to the vocal print feature using the pattern recognition model pre-established
To label information corresponding with the vocal print feature；
Memory module, for storing the label information is corresponding with the voiceprint into voice print database, so as to
Voiceprint is screened according to the label information in Application on Voiceprint Recognition.
Optionally, the processing module includes:
Analytical unit, for being analyzed using the pattern recognition model that pre-establishes the vocal print feature, obtain with
The corresponding pattern information of the vocal print feature；The pattern information includes language information, age information, gender information or native place letter
One of breath or a variety of combinations；
Generation unit, for generating label information corresponding with the vocal print feature according to the pattern information.
The embodiment of the invention also provides a kind of voice print databases to establish system, comprising:
Memory, for storing computer program；
Processor realizes the step of voice print database method for building up as described above when for executing the computer program
Suddenly.
The embodiment of the invention also provides a kind of method for recognizing sound-groove, comprising:
Voiceprint to be identified is obtained, extracts corresponding personal label information according to the voiceprint to be identified；
Foundation and the personal label information filter out corresponding more with the personal label information from voice print database
A voiceprint；The voice print database is the voice print database established according to voice print database method for building up as described above
Library；
The voiceprint to be identified is matched one by one with each voiceprint, is obtained and the sound to be identified
The corresponding target voiceprint of line information, and obtain identity information corresponding with the target voiceprint.
The embodiment of the present invention has also correspondingly provided a kind of voice print identification device, comprising:
Second extraction module extracts corresponding for obtaining voiceprint to be identified according to the voiceprint to be identified
Personal label information；
Screening module, for being filtered out and the personal label letter according to the personal label information from voice print database
Cease corresponding multiple voiceprints；The voice print database is to establish according to voice print database method for building up as described above
Voice print database；
Matching module is obtained for being matched the voiceprint to be identified one by one with each voiceprint
Target voiceprint corresponding with the voiceprint to be identified, and obtain identity corresponding with the target voiceprint and believe
Breath.
The embodiment of the invention also provides a kind of Voiceprint Recognition Systems, comprising:
Memory, for storing computer program；
Processor realizes the step of voice print database method for building up as described above when for executing the computer program
Suddenly.
The embodiment of the invention provides a kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system, packets
It includes: obtaining voiceprint, and extract the vocal print feature of voiceprint；Using the pattern recognition model pre-established to vocal print feature
Pattern match is carried out, label information corresponding with vocal print feature is obtained；It stores label information is corresponding with voiceprint to vocal print
In database, to be screened according to label information to voiceprint in Application on Voiceprint Recognition.As it can be seen that passing through the sound in the application
Include not only voiceprint in the voice print database that line database building method is established, further includes mark corresponding with voiceprint
Information is signed, so that the range of search of voiceprint can be reduced according to corresponding label information in subsequent Application on Voiceprint Recognition, from
And the recognition efficiency of Application on Voiceprint Recognition is improved, reduce Application on Voiceprint Recognition the time it takes.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to institute in the prior art and embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a kind of flow diagram of voice print database method for building up provided in an embodiment of the present invention；
Fig. 2 is the structural schematic diagram that a kind of voice print database provided in an embodiment of the present invention establishes device；
Fig. 3 is a kind of flow diagram of method for recognizing sound-groove provided in an embodiment of the present invention；
Fig. 4 is a kind of structural schematic diagram of voice print identification device provided in an embodiment of the present invention.
Specific embodiment
The embodiment of the invention provides a kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system, pass through
The voice print database established in the application is conducive to improve when carrying out Application on Voiceprint Recognition the recognition efficiency of Application on Voiceprint Recognition, reduction sound
Line identifies the time it takes.
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 is please referred to, Fig. 1 is a kind of flow diagram of voice print database method for building up provided in an embodiment of the present invention.
This method, comprising:
S110: voiceprint is obtained, and extracts the vocal print feature of voiceprint；
It should be noted that in the application when establishing voice print database, it can be in the acoustic information for acquiring corresponding personnel
Afterwards by handling to obtain its voiceprint accordingly, vocal print feature extraction then is carried out to the voiceprint again, is obtained corresponding
Vocal print feature；Wherein, extracted vocal print feature may include one in the sound intensity, loudness, pitch, pitch period or fundamental frequency
Kind or it is a variety of, it is, of course, also possible to include signal-to-noise ratio, humorous ratio of making an uproar, lock in phenomenon, amplitude around etc. other phonetic features, it is specific this
Application is without limitation.
Specifically, in physics, being averaged by the unit area perpendicular to Acoustic Wave Propagation direction in the unit time
Sound energy, the referred to as sound intensity；Loudness is a kind of subjective psychology amount, is the sound intensity degree that human subject feels, in general sound
The timing of voice frequency one, the sound intensity is stronger, and loudness is also bigger, but loudness is related with frequency, loudness when identical sound intensity frequency difference
It may be different；Pitch refers to the sound of various different height, the i.e. height of sound, one kind of sound essential characteristic；Pitch period and
Fundamental frequency is also one of vocal print feature feature.
S120: carrying out pattern match to vocal print feature using the pattern recognition model that pre-establishes, obtains and vocal print feature
Corresponding label information；
Specifically, the pattern recognition model in the application is preparatory according to preset algorithm and the progress of multiple history voiceprints
It establishes, establishment process are as follows:
Analyzing and training is carried out using preset algorithm pair vocal print feature corresponding with multiple history voiceprints, obtains mould
Formula identification model；Preset algorithm includes in languages recognizer, gender recognizer, age recognizer or native place recognizer
One or more combinations.
It is understood that acquiring a large amount of voiceprint (these voiceprints are as history voiceprint) in advance, lead to
It crosses after languages recognizer is analyzed and processed these history voiceprints, obtains languages pattern recognition model, pass through gender
After recognizer is analyzed and processed these history voiceprints, gender pattern recognition model is obtained, is identified and is calculated by the age
Method is analyzed and processed these history voiceprints, obtains age pattern identification model, by native place recognizer to here
History voiceprint is analyzed and processed, and obtains native place pattern recognition model, wherein languages recognizer, gender recognizer,
Age recognizer or native place recognizer are based on ivector algorithm.
Wherein, pattern match is carried out to vocal print feature using the pattern recognition model that pre-establishes in the application, obtain with
The process of the corresponding label information of vocal print feature, is specifically as follows:
Vocal print feature is analyzed using the pattern recognition model pre-established, obtains mode corresponding with vocal print feature
Information；Pattern information includes one of language information, age information, gender information or native place information or a variety of combinations；According to
Label information corresponding with vocal print feature is generated according to pattern information.
Specifically, language can be passed through when carrying out pattern match to the vocal print feature extracted using pattern recognition model
Kind pattern recognition model carries out analysis matching to vocal print feature, obtains language information；By gender pattern recognition model to vocal print
Feature carries out separation matching, obtains gender information；Analysis matching is carried out to vocal print feature by age pattern identification model, is obtained
Age information carries out analysis to vocal print feature by native place recognizer and matches to obtain ground native place information.That is, being obtained in the application
To after corresponding pattern information, label information can be generated according to specific pattern information, such as mould is being carried out to vocal print feature
After formula matching, obtained pattern information is age information, gender information and native place information, then can be according to the age information, property
Other information and native place information generate corresponding with voiceprint label information (such as: 45- male-Tianjin).
S130: it stores label information is corresponding with voiceprint into voice print database, so as to the foundation in Application on Voiceprint Recognition
Label information screens voiceprint.
Specifically, after obtaining label information corresponding with vocal print feature, the label information is corresponding with the voiceprint
It stores in voice print database, certainly, while the also identity information corresponding with the voiceprint stored, it can specifically incite somebody to action
In label information, identity information and identity information write-in same record, and by the record storage into voice print database.When
So, the above method is all made of for each voiceprint got and obtains corresponding label information, by label information-
Voiceprint-identity information record, and store into voice print database, to obtain being stored with the voice print database largely recorded
Library, to be used for subsequent Application on Voiceprint Recognition.When carrying out Application on Voiceprint Recognition according to the voice print database, can be believed by vocal print to be identified
Breath extracts corresponding personal label information, such as age-sex-native place, thus according to individual's label information to sound
Screened in line database, thus filter out with the consistent multiple label informations of individual's label information, and then obtain with often
The corresponding voiceprint of a label information, namely multiple voiceprints are filtered out, so that the matched range of voiceprint is reduced,
Again voiceprint to be identified is matched one by one with the multiple voiceprints filtered out, is determined consistent with the voiceprint
Target voiceprint, and then identity information corresponding with the target voiceprint is obtained, to greatly improve Application on Voiceprint Recognition effect
Rate.
In addition, it should also be noted that, in the application when establishing voice print database, it can also be in former voice print database
On the basis of establish, voiceprint gathered in advance and identity information corresponding with voiceprint are stored in legacy data library, can
To obtain voiceprint one by one from former voice print database by the method in the application, and voiceprint is handled one by one and is obtained
Label information corresponding with each voiceprint, and label information is written in record corresponding with corresponding voiceprint, from
And obtain new voice print database.
The embodiment of the invention provides a kind of voice print database method for building up, comprising: obtains voiceprint, and extracts vocal print
The vocal print feature of information；Pattern match is carried out to vocal print feature using the pattern recognition model pre-established, is obtained and vocal print spy
Levy corresponding label information；It stores label information is corresponding with voiceprint into voice print database, so as in Application on Voiceprint Recognition
Voiceprint is screened according to label information.As it can be seen that established by the voice print database method for building up in the application
Include not only voiceprint in voice print database, further include label information corresponding with voiceprint, to know in subsequent vocal print
The range of search of voiceprint can be reduced according to corresponding label information when other, to improve the identification effect of Application on Voiceprint Recognition
Rate reduces Application on Voiceprint Recognition the time it takes.
In the basic row of above-described embodiment, the embodiment of the present invention provides a kind of voice print database accordingly and establishes device,
Specifically referring to figure 2., which includes:
First extraction module 21 for obtaining voiceprint, and extracts the vocal print feature of voiceprint；
Processing module 22 is obtained for carrying out pattern match to vocal print feature using the pattern recognition model pre-established
Label information corresponding with vocal print feature；
Memory module 23, for storing label information is corresponding with voiceprint into voice print database, so as in vocal print
Voiceprint is screened according to label information when identification.
On the basis of a upper embodiment, further, processing module 22 includes:
Analytical unit is obtained and vocal print for being analyzed using the pattern recognition model that pre-establishes vocal print feature
The corresponding pattern information of feature；Pattern information include one of language information, age information, gender information or native place information or
A variety of combinations；
Generation unit, for generating label information corresponding with vocal print feature according to pattern information.
It should be noted that voice print database provided in the present embodiment, which establishes device, to be had and mentions in above-described embodiment
The identical beneficial effect of voice print database method for building up of confession, and to voice print database involved in this present embodiment
The introduction of method for building up, please refers to above-described embodiment, and this embodiment is not repeated.
Based on any of the above embodiments, the embodiment of the invention also provides a kind of voice print databases to establish system.
The system includes:
Memory, for storing computer program；
Processor is realized when for executing computer program such as the step of above-mentioned voice print database method for building up.
For example, processor and extracts the vocal print feature of voiceprint for realizing voiceprint is obtained；Using pre-establishing
Pattern recognition model to vocal print feature carry out pattern match, obtain label information corresponding with vocal print feature；By label information
It is corresponding with voiceprint to store into voice print database, to be sieved according to label information to voiceprint in Application on Voiceprint Recognition
Choosing.
Based on any of the above embodiments, the embodiment of the invention provides a kind of computer readable storage medium, meters
It is stored with computer program on calculation machine readable storage medium storing program for executing, such as above-mentioned voice print database is realized when computer program is executed by processor
The step of library method for building up.
Specifically, the computer readable storage medium may include: USB flash disk, mobile hard disk, read-only memory (Read-Only
Memory, ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. is various to deposit
Store up the medium of program code.
Based on any of the above embodiments, the embodiment of the invention also provides a kind of method for recognizing sound-groove, this method
It is corresponding with the voice print database method for building up in above-described embodiment, specifically referring to figure 3..This method comprises:
S310: obtaining voiceprint to be identified, extracts corresponding personal label information according to voiceprint to be identified；
S320: foundation and personal label information filter out multiple sound corresponding with personal label information from voice print database
Line information；Voice print database is according to the voice print database established such as above-mentioned voice print database method for building up；
S330: voiceprint to be identified is matched one by one with each voiceprint, is obtained and voiceprint to be identified
Corresponding target voiceprint, and obtain identity information corresponding with target voiceprint.
It should be noted that method for recognizing sound-groove provided in the present embodiment is applied to based on using in above-described embodiment
The voice print database that provided voice print database method for building up is established.
Specifically, identified to voiceprint to be identified, to determine identity corresponding with the voiceprint to be identified
When information, corresponding personal label information, such as age-sex-native place can be extracted by voiceprint to be identified,
To be screened according to individual's label information into voice print database, to filter out consistent with individual's label information
Multiple label informations, and then voiceprint corresponding with each label information is obtained, namely filter out multiple voiceprints, thus
It reduces the matched range of voiceprint, then voiceprint to be identified is carried out with the multiple voiceprints filtered out one by one
Match, determine with the consistent target voiceprint of the voiceprint, and then obtain corresponding with target voiceprint identity letter
Breath reduces identification process the time it takes to greatly improve Application on Voiceprint Recognition efficiency.The Application on Voiceprint Recognition provided in the present embodiment
Method can be used for the authentication of public security system, is conducive to the difficulty for mitigating criminal investigation work, can be also used for enterprises and institutions
In the work such as authentication, convenient for the personal management of relevant enterprise, it is of course also possible to for other specific fields, this Shen
Please without limitation.
On the basis of the above embodiments, the embodiment of the present invention has also correspondingly provided a kind of voice print identification device, specifically
Referring to figure 4..The device includes:
Second extraction module 41 extracts corresponding according to voiceprint to be identified for obtaining voiceprint to be identified
People's label information；
Screening module 42, for corresponding with personal label information according to being filtered out with personal label information from voice print database
Multiple voiceprints；Voice print database is according to the voice print database established such as above-mentioned voice print database method for building up；
Matching module 43, for voiceprint to be identified to be matched one by one with each voiceprint, obtain with wait know
The corresponding target voiceprint of other voiceprint, and obtain identity information corresponding with target voiceprint.
Voice print identification device provided in the present embodiment has and the method for recognizing sound-groove that provides in above-described embodiment
Identical beneficial effect, and the introduction to method for recognizing sound-groove involved in this present embodiment, please refer to above-mentioned implementation
Example, this embodiment is not repeated.
Based on any of the above embodiments, the embodiment of the invention also provides a kind of Voiceprint Recognition Systems, comprising:
Memory, for storing computer program；
Processor is realized when for executing computer program such as the step of above-mentioned voice print database method for building up.
For example, processor in the present embodiment is for realizing voiceprint to be identified is obtained, according to voiceprint to be identified
Extract corresponding personal label information；It is filtered out from voice print database and personal label information pair according to personal label information
The multiple voiceprints answered；Voice print database is according to the voice print database established such as above-mentioned voice print database method for building up；It will
Voiceprint to be identified is matched one by one with each voiceprint, obtains target vocal print letter corresponding with voiceprint to be identified
Breath, and obtain identity information corresponding with target voiceprint.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For device disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part
It is bright.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (10)
1. a kind of voice print database method for building up characterized by comprising
Voiceprint is obtained, and extracts the vocal print feature of the voiceprint；
Pattern match is carried out to the vocal print feature using the pattern recognition model pre-established, is obtained and the vocal print feature pair
The label information answered；
Store the label information is corresponding with the voiceprint into voice print database, so as in Application on Voiceprint Recognition according to institute
Label information is stated to screen voiceprint.
2. voice print database method for building up according to claim 1, which is characterized in that described using the mode pre-established
Identification model carries out pattern match to the vocal print feature, obtains the process packet of label information corresponding with the vocal print feature
It includes:
The vocal print feature is analyzed using the pattern recognition model pre-established, is obtained corresponding with the vocal print feature
Pattern information；The pattern information includes one of language information, age information, gender information or native place information or a variety of
Combination；
Label information corresponding with the vocal print feature is generated according to the pattern information.
3. voice print database method for building up according to claim 1 or 2, which is characterized in that the pattern recognition model
Method for building up includes:
Analyzing and training is carried out using preset algorithm pair vocal print feature corresponding with multiple history voiceprints, obtains the mould
Formula identification model；The preset algorithm includes that languages recognizer, gender recognizer, age recognizer or native place identification are calculated
One of method or a variety of combinations.
4. voice print database method for building up according to claim 3, which is characterized in that the vocal print feature include the sound intensity,
One of loudness, pitch, pitch period or fundamental frequency are a variety of.
5. a kind of voice print database establishes device characterized by comprising
First extraction module for obtaining voiceprint, and extracts the vocal print feature of the voiceprint；
Processing module, for carrying out pattern match to the vocal print feature using the pattern recognition model that pre-establishes, obtain with
The corresponding label information of the vocal print feature；
Memory module, for storing the label information is corresponding with the voiceprint into voice print database, so as in sound
Line screens voiceprint according to the label information when identifying.
6. voice print database according to claim 5 establishes device, which is characterized in that the processing module includes:
Analytical unit, for being analyzed using the pattern recognition model that pre-establishes the vocal print feature, obtain with it is described
The corresponding pattern information of vocal print feature；The pattern information includes in language information, age information, gender information or native place information
One or more combinations；
Generation unit, for generating label information corresponding with the vocal print feature according to the pattern information.
7. a kind of voice print database establishes system characterized by comprising
Memory, for storing computer program；
Processor realizes the voice print database as described in Claims 1-4 any one when for executing the computer program
The step of method for building up.
8. a kind of method for recognizing sound-groove characterized by comprising
Voiceprint to be identified is obtained, extracts corresponding personal label information according to the voiceprint to be identified；
Foundation and the personal label information filter out multiple sound corresponding with individual's label information from voice print database
Line information；The voice print database is to establish according to the voice print database method for building up as described in claim 1-4 any one
Voice print database；
The voiceprint to be identified is matched one by one with each voiceprint, obtains believing with the vocal print to be identified
Corresponding target voiceprint is ceased, and obtains identity information corresponding with the target voiceprint.
9. a kind of voice print identification device characterized by comprising
Second extraction module extracts corresponding personal for obtaining voiceprint to be identified according to the voiceprint to be identified
Label information；
Screening module, for being filtered out and the personal label information pair according to the personal label information from voice print database
The multiple voiceprints answered；The voice print database is to build according to the voice print database as described in claim 1-4 any one
The voice print database that cube method is established；
Matching module obtains and institute for being matched the voiceprint to be identified one by one with each voiceprint
The corresponding target voiceprint of voiceprint to be identified is stated, and obtains identity information corresponding with the target voiceprint.
10. a kind of Voiceprint Recognition System characterized by comprising
Memory, for storing computer program；
Processor realizes voice print database method for building up as claimed in claim 8 when for executing the computer program
Step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811089314.8A CN109036436A (en) | 2018-09-18 | 2018-09-18 | A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811089314.8A CN109036436A (en) | 2018-09-18 | 2018-09-18 | A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109036436A true CN109036436A (en) | 2018-12-18 |
Family
ID=64616830
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811089314.8A Pending CN109036436A (en) | 2018-09-18 | 2018-09-18 | A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109036436A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110797032A (en) * | 2020-01-06 | 2020-02-14 | 深圳中创华安科技有限公司 | Voiceprint database establishing method and voiceprint identification method |
| CN111326163A (en) * | 2020-04-15 | 2020-06-23 | 厦门快商通科技股份有限公司 | Voiceprint recognition method, device and equipment |
| CN111415669A (en) * | 2020-04-15 | 2020-07-14 | 厦门快商通科技股份有限公司 | Voiceprint model construction method, device and equipment |
| CN111444377A (en) * | 2020-04-15 | 2020-07-24 | 厦门快商通科技股份有限公司 | Voiceprint identification authentication method, device and equipment |
| CN111444376A (en) * | 2020-04-15 | 2020-07-24 | 厦门快商通科技股份有限公司 | Audio fingerprint identification method and device and equipment |
| CN111444375A (en) * | 2020-04-15 | 2020-07-24 | 厦门快商通科技股份有限公司 | Voiceprint identification verification method, device and equipment |
| CN111506763A (en) * | 2020-04-15 | 2020-08-07 | 厦门快商通科技股份有限公司 | Voiceprint database establishing method, device and equipment |
| CN111581426A (en) * | 2020-04-30 | 2020-08-25 | 厦门快商通科技股份有限公司 | Audio fingerprint matching method, device and equipment |
| CN111798857A (en) * | 2019-04-08 | 2020-10-20 | 北京嘀嘀无限科技发展有限公司 | Information identification method and device, electronic equipment and storage medium |
| CN112201233A (en) * | 2020-09-01 | 2021-01-08 | 沈澈 | Voice control method, system and device of intelligent household equipment and computer storage medium |
| CN112447178A (en) * | 2019-08-28 | 2021-03-05 | 北京声智科技有限公司 | Voiceprint retrieval method and device and electronic equipment |
| CN112466310A (en) * | 2020-10-15 | 2021-03-09 | 讯飞智元信息科技有限公司 | Deep learning voiceprint recognition method and device, electronic equipment and storage medium |
| CN112802482A (en) * | 2021-04-15 | 2021-05-14 | 北京远鉴信息技术有限公司 | Voiceprint serial-parallel identification method, individual soldier system and storage medium |
| CN113127673A (en) * | 2021-03-23 | 2021-07-16 | 上海掌数科技有限公司 | Voiceprint database construction method and data calling method thereof |
| CN113239229A (en) * | 2021-06-17 | 2021-08-10 | 张鹏涛 | Intelligent screening data processing method and system and cloud platform |
| CN113555022A (en) * | 2021-07-23 | 2021-10-26 | 平安科技（深圳）有限公司 | Voice-based same-person identification method, device, equipment and storage medium |
| CN113707154A (en) * | 2021-09-03 | 2021-11-26 | 上海瑾盛通信科技有限公司 | Model training method and device, electronic equipment and readable storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0197810A1 (en) * | 1985-03-06 | 1986-10-15 | Régis Vanoni | Method and device for recognizing an individual |
| CN1283843A (en) * | 1999-08-10 | 2001-02-14 | 国际商业机器公司 | Speech datas extraction |
| JP2006039777A (en) * | 2004-07-23 | 2006-02-09 | Sanyo Electric Co Ltd | Registration method and apparatus, and authentication method and apparatus |
| CN101447188A (en) * | 2007-11-27 | 2009-06-03 | 北京金运昭德数字科技有限公司 | Digital voice print identification system and validation and identification method |
| JP4311322B2 (en) * | 2004-09-28 | 2009-08-12 | ソニー株式会社 | Viewing content providing system and viewing content providing method |
| CN102802114A (en) * | 2012-06-20 | 2012-11-28 | 北京语言大学 | Method and system for screening seat by using voices |
| CN204791050U (en) * | 2015-04-01 | 2015-11-18 | 北京市商汤科技开发有限公司 | Authentication equipment |
| CN105244031A (en) * | 2015-10-26 | 2016-01-13 | 北京锐安科技有限公司 | Speaker identification method and device |
| CN105513597A (en) * | 2015-12-30 | 2016-04-20 | 百度在线网络技术（北京）有限公司 | Voiceprint authentication processing method and apparatus |
| CN107424615A (en) * | 2017-07-26 | 2017-12-01 | 北京捷通华声科技股份有限公司 | Method for recognizing sound-groove and device |
| CN107481720A (en) * | 2017-06-30 | 2017-12-15 | 百度在线网络技术（北京）有限公司 | A kind of explicit method for recognizing sound-groove and device |
| CN107492379A (en) * | 2017-06-30 | 2017-12-19 | 百度在线网络技术（北京）有限公司 | A kind of voice-print creation and register method and device |
-
2018
- 2018-09-18 CN CN201811089314.8A patent/CN109036436A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0197810A1 (en) * | 1985-03-06 | 1986-10-15 | Régis Vanoni | Method and device for recognizing an individual |
| CN1283843A (en) * | 1999-08-10 | 2001-02-14 | 国际商业机器公司 | Speech datas extraction |
| JP2006039777A (en) * | 2004-07-23 | 2006-02-09 | Sanyo Electric Co Ltd | Registration method and apparatus, and authentication method and apparatus |
| JP4311322B2 (en) * | 2004-09-28 | 2009-08-12 | ソニー株式会社 | Viewing content providing system and viewing content providing method |
| CN101447188A (en) * | 2007-11-27 | 2009-06-03 | 北京金运昭德数字科技有限公司 | Digital voice print identification system and validation and identification method |
| CN102802114A (en) * | 2012-06-20 | 2012-11-28 | 北京语言大学 | Method and system for screening seat by using voices |
| CN204791050U (en) * | 2015-04-01 | 2015-11-18 | 北京市商汤科技开发有限公司 | Authentication equipment |
| CN105244031A (en) * | 2015-10-26 | 2016-01-13 | 北京锐安科技有限公司 | Speaker identification method and device |
| CN105513597A (en) * | 2015-12-30 | 2016-04-20 | 百度在线网络技术（北京）有限公司 | Voiceprint authentication processing method and apparatus |
| CN107481720A (en) * | 2017-06-30 | 2017-12-15 | 百度在线网络技术（北京）有限公司 | A kind of explicit method for recognizing sound-groove and device |
| CN107492379A (en) * | 2017-06-30 | 2017-12-19 | 百度在线网络技术（北京）有限公司 | A kind of voice-print creation and register method and device |
| CN107424615A (en) * | 2017-07-26 | 2017-12-01 | 北京捷通华声科技股份有限公司 | Method for recognizing sound-groove and device |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798857A (en) * | 2019-04-08 | 2020-10-20 | 北京嘀嘀无限科技发展有限公司 | Information identification method and device, electronic equipment and storage medium |
| CN112447178A (en) * | 2019-08-28 | 2021-03-05 | 北京声智科技有限公司 | Voiceprint retrieval method and device and electronic equipment |
| CN110797032A (en) * | 2020-01-06 | 2020-02-14 | 深圳中创华安科技有限公司 | Voiceprint database establishing method and voiceprint identification method |
| CN111444376A (en) * | 2020-04-15 | 2020-07-24 | 厦门快商通科技股份有限公司 | Audio fingerprint identification method and device and equipment |
| CN111326163A (en) * | 2020-04-15 | 2020-06-23 | 厦门快商通科技股份有限公司 | Voiceprint recognition method, device and equipment |
| CN111444375A (en) * | 2020-04-15 | 2020-07-24 | 厦门快商通科技股份有限公司 | Voiceprint identification verification method, device and equipment |
| CN111506763A (en) * | 2020-04-15 | 2020-08-07 | 厦门快商通科技股份有限公司 | Voiceprint database establishing method, device and equipment |
| CN111444377A (en) * | 2020-04-15 | 2020-07-24 | 厦门快商通科技股份有限公司 | Voiceprint identification authentication method, device and equipment |
| CN111415669A (en) * | 2020-04-15 | 2020-07-14 | 厦门快商通科技股份有限公司 | Voiceprint model construction method, device and equipment |
| CN111581426A (en) * | 2020-04-30 | 2020-08-25 | 厦门快商通科技股份有限公司 | Audio fingerprint matching method, device and equipment |
| CN112201233A (en) * | 2020-09-01 | 2021-01-08 | 沈澈 | Voice control method, system and device of intelligent household equipment and computer storage medium |
| CN112466310A (en) * | 2020-10-15 | 2021-03-09 | 讯飞智元信息科技有限公司 | Deep learning voiceprint recognition method and device, electronic equipment and storage medium |
| CN113127673A (en) * | 2021-03-23 | 2021-07-16 | 上海掌数科技有限公司 | Voiceprint database construction method and data calling method thereof |
| CN113127673B (en) * | 2021-03-23 | 2022-07-22 | 上海掌数科技有限公司 | Method for constructing voiceprint database and data calling method thereof |
| CN112802482A (en) * | 2021-04-15 | 2021-05-14 | 北京远鉴信息技术有限公司 | Voiceprint serial-parallel identification method, individual soldier system and storage medium |
| CN113239229A (en) * | 2021-06-17 | 2021-08-10 | 张鹏涛 | Intelligent screening data processing method and system and cloud platform |
| CN113555022A (en) * | 2021-07-23 | 2021-10-26 | 平安科技（深圳）有限公司 | Voice-based same-person identification method, device, equipment and storage medium |
| CN113707154A (en) * | 2021-09-03 | 2021-11-26 | 上海瑾盛通信科技有限公司 | Model training method and device, electronic equipment and readable storage medium |
| CN113707154B (en) * | 2021-09-03 | 2023-11-10 | 上海瑾盛通信科技有限公司 | Model training method, device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109036436A (en) | A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system | |
| CN104835498B (en) | Method for recognizing sound-groove based on polymorphic type assemblage characteristic parameter | |
| CN105023573B (en) | It is detected using speech syllable/vowel/phone boundary of auditory attention clue | |
| CN107274916B (en) | Method and device for operating audio/video file based on voiceprint information | |
| CN109726624A (en) | Identity identifying method, terminal device and computer readable storage medium | |
| EP3206205A1 (en) | Voiceprint information management method and device as well as identity authentication method and system | |
| CN110459204A (en) | Audio recognition method, device, storage medium and electronic equipment | |
| CN110047490A (en) | Method for recognizing sound-groove, device, equipment and computer readable storage medium | |
| CN110222841A (en) | Neural network training method and device based on spacing loss function | |
| CN112712809B (en) | Voice detection method and device, electronic equipment and storage medium | |
| CN106294331A (en) | Audio information retrieval method and device | |
| CN112750442B (en) | Crested mill population ecological system monitoring system with wavelet transformation and method thereof | |
| CN106709804A (en) | Interactive wealth planning consulting robot system | |
| CN104538035A (en) | Speaker recognition method and system based on Fisher supervectors | |
| CN110136726A (en) | A kind of estimation method, device, system and the storage medium of voice gender | |
| CN108364656A (en) | A kind of feature extracting method and device for speech playback detection | |
| CN103871417A (en) | Specific continuous voice filtering method and device of mobile phone | |
| CN109766419A (en) | Products Show method, apparatus, equipment and storage medium based on speech analysis | |
| Ferragne et al. | Towards phonetic interpretability in deep learning applied to voice comparison | |
| CN110491368B (en) | Dialect background-based voice recognition method, device, computer equipment and storage medium | |
| CN109817223A (en) | Phoneme notation method and device based on audio-frequency fingerprint | |
| CN109657186A (en) | A kind of demographic method, system and relevant apparatus | |
| CN109545226A (en) | A kind of audio recognition method, equipment and computer readable storage medium | |
| CN110850419B (en) | Handwriting signature recognition method based on acoustics | |
| CN111326161B (en) | Voiceprint determining method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181218 |
|
| RJ01 | Rejection of invention patent application after publication |