CN110059059A - Batch screening technique, device, computer equipment and the storage medium of voice messaging - Google Patents
Batch screening technique, device, computer equipment and the storage medium of voice messaging Download PDFInfo
- Publication number
- CN110059059A CN110059059A CN201910197526.6A CN201910197526A CN110059059A CN 110059059 A CN110059059 A CN 110059059A CN 201910197526 A CN201910197526 A CN 201910197526A CN 110059059 A CN110059059 A CN 110059059A
- Authority
- CN
- China
- Prior art keywords
- voice messaging
- preset
- address
- readable text
- characteristic information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/14—Details of searching files based on file metadata
- G06F16/148—File search processing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/17—Details of further file system functions
- G06F16/172—Caching, prefetching or hoarding of files
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/61—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/16—Speech classification or search using artificial neural networks
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Library & Information Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Telephonic Communication Services (AREA)
Abstract
The embodiment of the invention discloses batch screening technique, device, computer equipment and the storage mediums of a kind of voice messaging, if wherein obtaining the import folders address of the file where preset training set the method includes receiving message processing directives；Preset threshold and preset first export folders address, the second export folders address are determined according to the message processing directives；The import folders address is read to obtain all voice messagings to be processed；Preset voice screening script is called to extract the characteristic information of each voice messaging to be processed respectively；All characteristic informations are successively read to judge whether it matches with preset threshold；If so, storing voice messaging to be measured corresponding to this feature information into the second export folders corresponding to the second export folders address to be used for batch signatures.The present invention can efficiently and accurately realize the unified screening to multiple voice messagings to be processed in training set, and reduce the mistake of screening process.
Description
Technical field
The present invention relates to data processing field more particularly to batch screening technique, device, the computers of a kind of voice messaging
Equipment and storage medium.
Background technique
It usually requires to collect or acquire a large amount of voice messagings from various channels in speech recognition project, and utilizes these languages
Message breath is trained neural network as the training sample in training set, to obtain accordingly for carrying out the language of feature
The identification model of sound identification.And it is accurate in order to ensure the smooth and acquired identification model of the training process of neural network
Property, it usually needs the pre-processing before being trained to acquired voice messaging, such as the screening of effective voice messaging, and it is real
Now need progressive alternate that could complete the pretreatment work of a large amount of voice messaging, but the process factor of iteration processing
It is big according to amount, it is very easy to operation error occur, causes the problem of voice messaging screening inaccuracy.
Summary of the invention
The embodiment of the present invention provides batch screening technique, device, computer equipment and the storage medium of a kind of voice messaging,
It can efficiently and accurately realize the unified screening to multiple voice messagings to be processed in training set, and reduce the mistake of screening process
Accidentally.
In a first aspect, the embodiment of the invention provides a kind of batch screening techniques of voice messaging, this method comprises:
If receiving message processing directives, the address of the file where preset training set is obtained, and the address is made
For import folders address, the training set includes multiple voice messagings to be processed；
Preset threshold and preset first export folders address, the second output are determined according to the message processing directives
Folder address, wherein first export folders address is the address that the first export folders is saved, and described first is defeated
File includes multiple readable text files out, and second export folders address is the ground that the second export folders is saved
Location；
The import folders address is read to obtain all voice messagings to be processed；
Call preset voice screening script to extract the characteristic information of each voice messaging to be processed respectively, and will be each
The characteristic information of voice messaging to be processed is respectively written into different readable text files；
The characteristic information in all readable text files in first export folders is successively read to judge
State whether the characteristic information in readable text file matches with preset threshold；
It is if the characteristic information in the readable text file matches with preset threshold, the readable text file institute is right
The voice messaging to be measured answered is stored into second export folders for batch signatures.
Second aspect, the embodiment of the invention also provides a kind of batch screening plant of voice messaging, which includes using
In the unit for executing the above method.
The third aspect, the embodiment of the invention also provides a kind of computer equipments comprising memory and processor, it is described
Computer program is stored on memory, the processor realizes the above method when executing the computer program.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage medium, the storage medium storage
There is computer program, the computer program can realize the above method when being executed by a processor.
The embodiment of the invention provides a kind of batch screening technique of voice messaging, device, computer equipment and storages to be situated between
Matter.Wherein, which comprises if receiving message processing directives, the address of the file where preset training set is obtained,
And using the address as import folders address, the training set includes multiple voice messagings to be processed；At the information
Reason, which instructs, determines preset threshold and preset first export folders address, the second export folders address；It reads described defeated
Enter folder address to obtain all voice messagings to be processed；Call preset voice screening script with extract respectively each to
The characteristic information of voice messaging is handled, and the characteristic information of each voice messaging to be processed is respectively written into different readable texts
In file；It is described to judge to be successively read the characteristic information in all readable text files in first export folders
Whether the characteristic information in readable text file matches with preset threshold；If characteristic information in the readable text file with
Preset threshold matches, then stores voice messaging to be measured corresponding to the readable text file to second export folders
In be used for batch signatures.The embodiment of the present invention can efficiently and accurately be realized in training set by above-mentioned batch processing
The unified screening of multiple voice messagings to be processed, and the mistake of screening process is reduced, in order to accurately realize neural network
Training.
Detailed description of the invention
Technical solution in order to illustrate the embodiments of the present invention more clearly, below will be to needed in embodiment description
Attached drawing is briefly described, it should be apparent that, drawings in the following description are some embodiments of the invention, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of the batch screening technique of voice messaging provided in an embodiment of the present invention；
Fig. 2 is a kind of sub-process schematic diagram of the batch screening technique of voice messaging provided in an embodiment of the present invention；
Fig. 3 is a kind of sub-process schematic diagram of the batch screening technique of voice messaging provided in an embodiment of the present invention；
Fig. 4 be another embodiment of the present invention provides a kind of voice messaging batch screening technique flow diagram；
Fig. 5 is a kind of schematic block diagram of the batch screening plant of voice messaging provided in an embodiment of the present invention；
Fig. 6 is a kind of signal of the information determination unit of the batch screening plant of voice messaging provided in an embodiment of the present invention
Property block diagram；
Fig. 7 is a kind of signal of the information judging unit of the batch screening plant of voice messaging provided in an embodiment of the present invention
Property block diagram；
Fig. 8 be another embodiment of the present invention provides a kind of voice messaging batch screening plant schematic block diagram；
Fig. 9 is a kind of computer equipment structure composition schematic diagram provided in an embodiment of the present 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.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
Referring to Fig. 1, Fig. 1 is a kind of exemplary flow of the batch screening technique of voice messaging provided by the embodiments of the present application
Figure.The batch screening technique of the voice messaging is applied in management server.The management server management server passes through training
Before collection is trained neural network, batch pretreatment is carried out to the voice messaging to be processed in the training set got, such as
The voice messaging to be processed of damage, too short voice messaging to be processed are rejected from training set, it can by above-mentioned batch processing
It efficiently and accurately realizes the unified screening to multiple voice messagings to be processed in training set, and reduces the mistake of screening process,
In order to accurately realize the training of neural network.As shown in Figure 1, the step of this method includes step S101~S104.
Step S101 obtains the address of the file where preset training set, and will if receiving message processing directives
As import folders address, the training set includes multiple voice messagings to be processed for the address.
In the present embodiment, it in order to be trained neural network to obtaining corresponding speech recognition modeling, needs pair
Voice messaging to be processed in the training set got carries out the pretreatment of batch, meets wanting for trained neural network to reach
It asks, improves the precision for the speech recognition modeling that training obtains.And training set can be it is pre-set, it can from each energy
It enough carries out collection voice messaging in the application program of voice messaging acquisition to be stored, can also be through different recording personnel
Recording is carried out to obtain voice messaging, the voice messaging being stored in training set at this time is voice messaging to be processed.When
Management server receives Client-initiated message processing directives, then then obtaining the file where pre-set training set
Address, and using the address as import folders address, in order to which user is accurately located import folders, i.e. input file
Training set is stored in folder.
Step S102, with determining preset threshold and preset first export folders according to the message processing directives
Location, the second export folders address, wherein first export folders address is the ground that the first export folders is saved
Location, first export folders include multiple readable text files, and second export folders address is the second output text
The address that part double-layered quilt saves.
It in the present embodiment, may include pre-set preset threshold and preset in the message processing directives
One export folders address, the second export folders address, in order to which management server is after receiving message processing directives,
By analyzing the message processing directives to obtain parameter needed for audio screening process, these parameters may include above-mentioned
Preset threshold and preset first export folders address, the second export folders address.Meanwhile first output file
The address that folder address is saved as the first export folders is corresponding first export folders, first output file
Folder may include having multiple readable text files, and readable text file herein can be the empty text file of document name,
It is also possible to prestore the text file of enough memory spaces, i.e., the storage that readable File can be used for carrying out data is protected
It stays.Second export folders address is the address that the second export folders is saved, and is corresponding second export folders.
Wherein, audio is screened and is mainly sieved according to the characteristic information of of voice messaging to be processed itself
Choosing, therefore need to preset preset threshold, whether the characteristic information that voice messaging to be processed is defined by preset threshold meets
It is required that and the satisfactory voice messaging to be processed is stored to the second output text corresponding to the second export folders address
In part folder.For example, preset threshold can be the relevant threshold values of audio duration with voice messaging, adopting with voice messaging can be
The relevant threshold values of number of samples can also be the relevant threshold values with the zoom factor of voice messaging, be also possible to voice messaging
Relevant threshold values of maximum amplitude value etc..In addition, the first export folders corresponding to the first export folders address can be used for
Store intermediate file.
In one embodiment, as shown in Fig. 2, the step S102 may include step S201~S202.
Step S201 parses the message processing directives to obtain corresponding presupposed information.
Wherein, the message processing directives include the pre-set much information of user, in order to which management server exists
After obtaining the message processing directives, corresponding audio screening is carried out according to pre-set much information.
Step S202 determines preset threshold and preset first export folders address, according to the presupposed information
Two export folders addresses.
Wherein, in order to realize the accurate screening of voice messaging, management server can be determined according to the presupposed information exist
The parameter needed in audio screening process, such as preset threshold and preset first export folders address, the second output file
Press from both sides address.For example, the preset threshold can be the relevant preset duration threshold values with the audio duration of voice messaging, can also be
The relevant default sampling number threshold values with the sampled point of voice messaging.
Step S103 reads the import folders address to obtain all voice messagings to be processed.
In the present embodiment, management server can read the import folders address, and according to the import folders
Address determines corresponding import folders, so that all voice messagings to be processed in corresponding import folders are obtained, with
Convenient for carrying out batch processing to all voice messagings to be processed.
Step S104 calls preset voice screening script to extract the feature letter of each voice messaging to be processed respectively
Breath, and the characteristic information of each voice messaging to be processed is respectively written into different readable text files.
In the present embodiment, preset format conversion script refers to and pre-set can be screened to voice messaging
Script, such as preset voice screening script can be SOX script, can also be other for carrying out audio screening certainly
Script, program or function etc..It, can batch after management server executes the preset audio screening script of calling by Python
Extract the characteristic information of each voice messaging to be processed in ground.It may include wherein voice messaging about the characteristic information of voice messaging
The information such as audio duration, sampling number, zoom factor and maximum amplitude value.For the ease of being carried out to voice messaging to be processed
Specific analysis, the characteristic information of each voice messaging to be processed can be stored into a corresponding readable text file into
Row record, under normal circumstances, the corresponding different readable text file of different voice messagings to be processed.As optional,
The readable text file can be TXT file, naturally it is also possible to be other text files convenient for read-write, such as word file.
In addition, for the ease of unified management, all readable text files can be stored in preset first output file
It presss from both sides in the first export folders corresponding to address, in order to which management server is called the operation such as extraction as needed.
Step S105, the characteristic information being successively read in all readable text files in first export folders
To judge whether the characteristic information in the readable text file matches with preset threshold.
In the present embodiment, management server can be successively read all machine readable texts in the first export folders herein
Characteristic information in part, and acquired characteristic information is gone to match with preset threshold, so that it is determined that meeting preset threshold
The voice messaging for the requirement defined.
In one embodiment, as shown in figure 3, the preset threshold includes preset duration threshold values, the characteristic information includes
Audio duration, the step S105 may include step S301~S303.
Step S301, when the audio being successively read in all readable text files in first export folders
It is long.
Wherein, management server can be successively from the sound read in all readable text files in the first export folders
Frequency duration, each corresponding voice messaging to be processed of readable text file, therefore knowing that management server is extracted should be
The audio duration of each voice messaging to be processed.
Step S302, judges whether the audio duration in the readable text file is greater than or equal to preset duration threshold values.
Wherein, when the audio duration of voice messaging is less than preset duration threshold values, the voice messaging may be indicated in training
Good effect can not be played during neural network, to guarantee training result, when can remain larger than or be equal to default
The voice messaging of long threshold values.So when need that batch is gone to judge that audio duration in the readable text file is greater than or equal to it is pre-
If duration threshold values.The preset duration threshold values can carry out equipment according to the actual demand situation of user, in the present embodiment, not
It limits.
Step S303 determines if the audio duration in the readable text file is greater than or equal to preset duration threshold values
Characteristic information in the readable text file matches with preset threshold.
Wherein, when the audio duration in readable text file is greater than or equal to preset duration threshold values, then then can be determined that
Characteristic information in the readable text file is matched with preset threshold, then shows that the readable text file is corresponding at this time
Voice messaging to be measured be effective voice messaging.
In addition, the step S105 may include:
Step S303a can described in judgement if the audio duration in the readable text file is less than preset duration threshold values
The characteristic information read in text file is not matched that with preset threshold.When the audio duration in the readable text file is less than in advance
If duration threshold values when, need to screen out voice messaging to be processed corresponding to the readable text file.
As further embodiment, can also include: before the step S303
Step S304 is successively read if the audio duration in the readable text file is greater than or equal to preset duration threshold values
Take the sampling number in all readable text files in first export folders.
Wherein, after the audio duration of voice messaging to be measured meets certain require, in order to further determine voice to be measured
Whether information is effective information, it is also necessary to analyze from sampling number voice messaging to be measured, therefore need successively to obtain
Take the sampling number in all readable text files.
Step S305 judges that the sampling number in the readable text file is greater than or equal to default sampling number.
Wherein, in order to ensure voice messaging to be measured is relatively sharp in playing process, the voice to be measured of selection is needed at this time
The sampling number of information needs to be greater than or equal to default sampling number, which can carry out according to the demand of user
Corresponding setting, in the present embodiment and without limitation.
Specifically, executing institute if the sampling number in the readable text file is greater than or equal to default sampling number
State the step of characteristic information determined in the readable text file matches with preset threshold, i.e. execution step S303.Wherein,
If being greater than or equal to default sampling number using points in the readable text file, show that the readable text file institute is right
The voice messaging to be processed answered is effective voice messaging, therefore can be determined that the characteristic information in the readable text file and pre-
What if threshold values matched.
In addition, executing the judgement institute if the sampling number in the readable text file is less than default sampling number
The step of characteristic information and preset threshold in readable text file do not match that is stated, S303a is thened follow the steps.Wherein, work as institute
When stating the audio duration in readable text file less than preset duration threshold values, then show that voice messaging to be measured at this time is not to be inconsistent
Requirement is closed, needs to screen out voice messaging to be processed corresponding to the readable text file.
Step S106, if the characteristic information in the readable text file matches with preset threshold, by the machine readable text
Voice messaging to be measured corresponding to this document is stored into second export folders for batch signatures.
In the present embodiment, to call the voice messaging for having carried out audio screening convenient for management server, basis is needed
Second export folders address determines the position of the second export folders, and characteristic information obtained from being screened in batches and pre-
If the voice messaging to be measured that threshold values matches is stored into the second export folders.
As further embodiment, the message processing directives include the 4th export folders address, and the method is also
May include:
Step S107, it is if the characteristic information in the readable text file is not matched that with preset threshold, this is readable
Voice messaging to be measured corresponding to text file is stored to the 4th output file corresponding to the 4th export folders address
In folder.
In the present embodiment, if the characteristic information in the readable text file is not matched that with preset threshold, show
Voice messaging to be processed corresponding to the readable text file be it is undesirable, can be according to described for the ease of management
Four export folders addresses determine its corresponding 4th export folders, and by voice to be measured corresponding to the readable text file
Information is stored into the 4th export folders.
To sum up, the present embodiment can be realized efficiently and accurately by above-mentioned batch processing to multiple wait locate in training set
The unified screening of voice messaging is managed, and reduces the mistake of screening process, in order to accurately realize the training of neural network.
Referring to Fig. 4, Fig. 4 be another embodiment of the present invention provides a kind of voice messaging batch screening technique signal
Flow chart.As shown in figure 4, the step of this method includes step S401~S404.Wherein with the step S101- in above-described embodiment
The relevant explanation of S106 similar step and it is described in detail that details are not described herein, the following detailed description of to be increased in the present embodiment
The step of adding.
Step S401 obtains the address of the file where preset training set, and will if receiving message processing directives
As import folders address, the training set includes multiple voice messagings to be processed for the address.
Step S402, with determining preset threshold and preset first export folders according to the message processing directives
Location, the second export folders address, wherein first export folders address is the ground that the first export folders is saved
Location, first export folders include multiple readable text files, and second export folders address is the second output text
The address that part double-layered quilt saves.
Step S403 reads the import folders address to obtain all voice messagings to be processed.
Step S403a, the feature letter being successively read in all readable text files in first export folders
Cease the audio format to determine each voice messaging to be processed respectively.
In the present embodiment, in order to further assure that management server preferably call preset voice screening script with
The characteristic information of each voice messaging to be processed of batch extracting, it is also necessary to carry out the audio format of each audio-frequency information to be processed
Unified conversion process.Wherein, audio format refer to specifically can be may include AIFF, MPEG, MP3, MIDI, WMA,
The formats such as FLAC, APE, AMR, WAV, management server are all readable in first export folders by being successively read
Characteristic information in text file can determine the audio format of each voice messaging to be processed in batches.
Step S403b is kept described to be processed if the audio format of the voice messaging to be processed is preset audio format
The audio format of voice messaging is constant.
Wherein, when the audio format of the voice messaging to be processed is preset audio format, then show at this time to be processed
Voice messaging does not need to be handled, it can keeps the audio format of the voice messaging to be processed constant.For example, when default
Audio format is WAV format, and when the audio format of voice messaging to be processed is also WAV format, then it does not need to carry out audio lattice
Formula conversion.
Step S403c, if the audio format of the voice messaging to be processed is not preset audio format, according to preset sound
The audio format of the voice messaging to be processed is converted to preset audio format by frequency format transformation rule.
It wherein, is that basis is needed to preset when the audio format of voice messaging to be processed is not preset audio format
Audio format transformation rule audio format conversion is carried out to it.For example, the preset audio format transformation rule can be
Audio format conversion is carried out to voice messaging to be processed by Ffmpeg script.When preset audio format is WAV format, and wait locate
When the audio format for managing voice messaging is also MP3 format, the audio format by the voice to be processed is needed to convert.
Step S404 calls preset voice screening script to extract the feature letter of each voice messaging to be processed respectively
Breath, and the characteristic information of each voice messaging to be processed is respectively written into different readable text files.
Step S405, the characteristic information being successively read in all readable text files in first export folders
To judge whether the characteristic information in the readable text file matches with preset threshold.
Step S406, if the characteristic information in the readable text file matches with preset threshold, by the machine readable text
Voice messaging to be measured corresponding to this document is stored into second export folders for batch signatures.
As further embodiment, the message processing directives include preset third export folders address, described
Method can with the following steps are included:
Step S407, the characteristic information being successively read in all readable text files in first export folders
To judge whether the type of the characteristic information in each readable text file matches with the type of preset characteristic information respectively.
Wherein, according to the type of the type of the characteristic information of voice messaging to be processed and preset characteristic information whether
Match, so that it may judge whether it is effective voice messaging, if such as characteristic information need include voice messaging audio duration,
Sampling number, zoom factor and maximum amplitude value, and the preservation in readable text file only only has audio duration and adopts
Number of samples, then then the corresponding voice messaging to be processed of the readable text part is invalid information.And in readable text file
Save include audio duration, sampling number, zoom factor and maximum amplitude value, then show the readable text part it is corresponding to
Processing voice messaging is effective information.
Step S408, if the type of the characteristic information in the readable text file and the type of preset characteristic information are not
Match, determine that voice messaging to be measured is invalid voice information corresponding to the readable text file, and by the voice to be measured
Information is stored into third export folders corresponding to third export folders address.
Wherein, in order to further distinguish the property of voice messaging to be measured, it can will be determined as the to be measured of invalid voice information
Voice messaging is stored into third export folders corresponding to third export folders address.
As further embodiment, the message processing directives include the 4th export folders address, and the method is also
May include:
Step S409, it is if the characteristic information in the readable text file is not matched that with preset threshold, this is readable
Voice messaging to be measured corresponding to text file is stored to the 4th output file corresponding to the 4th export folders address
In folder.
Those having ordinary skill in the art is understood that realize all or part of the process in above-described embodiment method, is that can lead to
Computer program is crossed to instruct relevant hardware and complete, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) etc..
Referring to Fig. 5, a kind of corresponding above-mentioned batch screening technique of voice messaging, the embodiment of the present invention also propose a kind of language
The batch screening plant of message breath, the device 100 include: address acquisition unit 101, information determination unit 102, information reading list
Member 103, feature extraction unit 104, information judging unit 105 and the first storage unit 106.
The address acquisition unit 101, if obtaining the text where preset training set for receiving message processing directives
The address of part folder, and using the address as import folders address, the training set includes multiple voice messagings to be processed.
In the present embodiment, it in order to be trained neural network to obtaining corresponding speech recognition modeling, needs pair
Voice messaging to be processed in the training set got carries out the pretreatment of batch, meets wanting for trained neural network to reach
It asks, improves the precision for the speech recognition modeling that training obtains.And training set can be it is pre-set, it can from each energy
It enough carries out collection voice messaging in the application program of voice messaging acquisition to be stored, can also be through different recording personnel
Recording is carried out to obtain voice messaging, the voice messaging being stored in training set at this time is voice messaging to be processed.When
Management server receives Client-initiated message processing directives, then then obtaining the file where pre-set training set
Address, and using the address as import folders address, in order to which user is accurately located import folders, i.e. input file
Training set is stored in folder.
The information determination unit 102, for determining preset threshold and preset according to the message processing directives
One export folders address, the second export folders address, wherein first export folders address is the first output file
The address that double-layered quilt saves, first export folders includes multiple readable text files, second export folders address
The address being saved for the second export folders.
It in the present embodiment, may include pre-set preset threshold and preset in the message processing directives
One export folders address, the second export folders address, in order to which management server is after receiving message processing directives,
By analyzing the message processing directives to obtain parameter needed for audio screening process, these parameters may include above-mentioned
Preset threshold and preset first export folders address, the second export folders address.Meanwhile first output file
The address that folder address is saved as the first export folders is corresponding first export folders, first output file
Folder may include having multiple readable text files, and readable text file herein can be the empty text file of document name,
It is also possible to prestore the text file of enough memory spaces, i.e., the storage that readable File can be used for carrying out data is protected
It stays.Second export folders address is the address that the second export folders is saved, and is corresponding second export folders.
Wherein, audio is screened and is mainly sieved according to the characteristic information of of voice messaging to be processed itself
Choosing, therefore need to preset preset threshold, whether the characteristic information that voice messaging to be processed is defined by preset threshold meets
It is required that and the satisfactory voice messaging to be processed is stored to the second output text corresponding to the second export folders address
In part folder.For example, preset threshold can be the relevant threshold values of audio duration with voice messaging, adopting with voice messaging can be
The relevant threshold values of number of samples can also be the relevant threshold values with the zoom factor of voice messaging, be also possible to voice messaging
Relevant threshold values of maximum amplitude value etc..In addition, the first export folders corresponding to the first export folders address can be used for
Store intermediate file.
In one embodiment, as shown in fig. 6, the information determination unit 102 may include instruction resolution unit 201 and
Information extraction unit 202.
Described instruction resolution unit 201, for parsing the message processing directives to obtain corresponding presupposed information.
Wherein, the message processing directives include the pre-set much information of user, in order to which management server exists
After obtaining the message processing directives, corresponding audio screening is carried out according to pre-set much information.
The information extraction unit 202, for determining preset threshold and preset first defeated according to the presupposed information
Folder address, the second export folders address out.
Wherein, in order to realize the accurate screening of voice messaging, management server can be determined according to the presupposed information exist
The parameter needed in audio screening process, such as preset threshold and preset first export folders address, the second output file
Press from both sides address.For example, the preset threshold can be the relevant preset duration threshold values with the audio duration of voice messaging, can also be
The relevant default sampling number threshold values with the sampled point of voice messaging.
The Information reading unit 103, for reading the import folders address to obtain all voices to be processed
Information.
In the present embodiment, management server can read the import folders address, and according to the import folders
Address determines corresponding import folders, so that all voice messagings to be processed in corresponding import folders are obtained, with
Convenient for carrying out batch processing to all voice messagings to be processed.
The feature extraction unit 104, for calling preset voice screening script to extract each language to be processed respectively
The characteristic information of message breath, and the characteristic information of each voice messaging to be processed is respectively written into different readable text files
In.
In the present embodiment, preset format conversion script refers to and pre-set can be screened to voice messaging
Script, such as preset voice screening script can be SOX script, can also be other for carrying out audio screening certainly
Script, program or function etc..It, can batch after management server executes the preset audio screening script of calling by Python
Extract the characteristic information of each voice messaging to be processed in ground.It may include wherein voice messaging about the characteristic information of voice messaging
The information such as audio duration, sampling number, zoom factor and maximum amplitude value.For the ease of being carried out to voice messaging to be processed
Specific analysis, the characteristic information of each voice messaging to be processed can be stored into a corresponding readable text file into
Row record, under normal circumstances, the corresponding different readable text file of different voice messagings to be processed.As optional,
The readable text file can be TXT file, naturally it is also possible to be other text files convenient for read-write, such as word file.
In addition, for the ease of unified management, all readable text files can be stored in preset first output file
It presss from both sides in the first export folders corresponding to address, in order to which management server is called the operation such as extraction as needed.
The information judging unit 105, all readable texts for being successively read in first export folders
Characteristic information in file is to judge whether the characteristic information in the readable text file matches with preset threshold.
In the present embodiment, management server can be successively read all machine readable texts in the first export folders herein
Characteristic information in part, and acquired characteristic information is gone to match with preset threshold, so that it is determined that meeting preset threshold
The voice messaging for the requirement defined.
In one embodiment, as shown in fig. 7, the preset threshold includes preset duration threshold values, the characteristic information includes
Audio duration, the information judging unit 105 may include duration reading unit 301, duration judging unit 302 and first sentence
Order member 303.
The duration reading unit 301, all readable texts for being successively read in first export folders
Audio duration in file.
Wherein, management server can be successively from the sound read in all readable text files in the first export folders
Frequency duration, each corresponding voice messaging to be processed of readable text file, therefore knowing that management server is extracted should be
The audio duration of each voice messaging to be processed.
The duration judging unit 302, for judging whether the audio duration in the readable text file is greater than or waits
In preset duration threshold values.
Wherein, when the audio duration of voice messaging is less than preset duration threshold values, the voice messaging may be indicated in training
Good effect can not be played during neural network, to guarantee training result, when can remain larger than or be equal to default
The voice messaging of long threshold values.So when need that batch is gone to judge that audio duration in the readable text file is greater than or equal to it is pre-
If duration threshold values.The preset duration threshold values can carry out equipment according to the actual demand situation of user, in the present embodiment, not
It limits.
First judging unit 303, if being greater than or equal to for the audio duration in the readable text file default
Duration threshold values then determines that the characteristic information in the readable text file matches with preset threshold.
Wherein, when the audio duration in readable text file is greater than or equal to preset duration threshold values, then then can be determined that
Characteristic information in the readable text file is matched with preset threshold, then shows that the readable text file is corresponding at this time
Voice messaging to be measured be effective voice messaging.
In addition, the information judging unit 105 may include:
Second judging unit 303a, if being less than preset duration threshold values for the audio duration in the readable text file,
Then determine that the characteristic information in the readable text file is not matched that with preset threshold.Sound in the readable text file
When frequency duration is less than preset duration threshold values, need to sieve voice messaging to be processed corresponding to the readable text file
It removes.
As further embodiment, can also include: before first judging unit 303
Numerical value reading unit 304, if being greater than or equal to preset duration for the audio duration in the readable text file
Threshold values, the sampling number being successively read in all readable text files in first export folders.
Wherein, after the audio duration of voice messaging to be measured meets certain require, in order to further determine voice to be measured
Whether information is effective information, it is also necessary to analyze from sampling number voice messaging to be measured, therefore need successively to obtain
Take the sampling number in all readable text files.
Numerical value judging unit 305, for judging that the sampling number in the readable text file is greater than or equal to default adopt
Number of samples.
Wherein, in order to ensure voice messaging to be measured is relatively sharp in playing process, the voice to be measured of selection is needed at this time
The sampling number of information needs to be greater than or equal to default sampling number, which can carry out according to the demand of user
Corresponding setting, in the present embodiment and without limitation.
Specifically, in one embodiment, if first judging unit 303 is also used to adopting in the readable text file
Number of samples is greater than or equal to default sampling number, determines characteristic information and preset threshold phase in the readable text file
Match.Wherein, if being greater than or equal to default sampling number using points in the readable text file, show the readable text
Voice messaging to be processed corresponding to file is effective voice messaging, therefore can be determined that the feature in the readable text file
What information and preset threshold matched.
In addition, in one embodiment, if the second judging unit 303a is also used to adopting in the readable text file
Number of samples is less than default sampling number, determines that the characteristic information in the readable text file is not matched that with preset threshold.Its
In, when the audio duration in the readable text file is less than preset duration threshold values, then show voice to be measured letter at this time
Breath be it is undesirable, need to screen out voice messaging to be processed corresponding to the readable text file.
First storage unit 106, if for characteristic information and preset threshold phase in the readable text file
Match, then stores voice messaging to be measured corresponding to the readable text file into second export folders to be used for batch
Output.
In the present embodiment, to call the voice messaging for having carried out audio screening convenient for management server, basis is needed
Second export folders address determines the position of the second export folders, and characteristic information obtained from being screened in batches and pre-
If the voice messaging to be measured that threshold values matches is stored into the second export folders.
As further embodiment, the message processing directives include the 4th export folders address, described device 100
Can also include:
Second storage unit 107, if the characteristic information in the readable text file is not matched that with preset threshold,
Then voice messaging to be measured corresponding to the readable text file is stored to corresponding to the 4th export folders address
In four export folders.
In the present embodiment, if the characteristic information in the readable text file is not matched that with preset threshold, show
Voice messaging to be processed corresponding to the readable text file be it is undesirable, can be according to described for the ease of management
Four export folders addresses determine its corresponding 4th export folders, and by voice to be measured corresponding to the readable text file
Information is stored into the 4th export folders.
Referring to Fig. 8, a kind of corresponding above-mentioned batch screening technique of voice messaging, another embodiment of the present invention also propose one
The batch screening plant of kind voice messaging, the device 400 include: address acquisition unit 401, information determination unit 402, information reading
Take unit 403, format determination unit 403a, format holding unit 403b, format conversion unit 403c, feature extraction unit 404,
Information judging unit 405 and the first storage unit 406.
Address acquisition unit 401, if obtaining the file where preset training set for receiving message processing directives
Address, and using the address as import folders address, the training set includes multiple voice messagings to be processed.
Information determination unit 402, for determining preset threshold and preset first defeated according to the message processing directives
Folder address, the second export folders address out, wherein first export folders address is the first output file double-layered quilt
The address of preservation, first export folders include multiple readable text files, and second export folders address is the
The address that two export folders are saved.
Information reading unit 403, for reading the import folders address to obtain all voice messagings to be processed.
Format determination unit 403a, all readable text files for being successively read in first export folders
In characteristic information to determine the audio format of each voice messaging to be processed respectively.
In the present embodiment, in order to further assure that management server preferably call preset voice screening script with
The characteristic information of each voice messaging to be processed of batch extracting, it is also necessary to carry out the audio format of each audio-frequency information to be processed
Unified conversion process.Wherein, audio format refer to specifically can be may include AIFF, MPEG, MP3, MIDI, WMA,
The formats such as FLAC, APE, AMR, WAV, management server are all readable in first export folders by being successively read
Characteristic information in text file can determine the audio format of each voice messaging to be processed in batches.
Format holding unit 403b is protected if the audio format for the voice messaging to be processed is preset audio format
The audio format for holding the voice messaging to be processed is constant.
Wherein, when the audio format of the voice messaging to be processed is preset audio format, then show at this time to be processed
Voice messaging does not need to be handled, it can keeps the audio format of the voice messaging to be processed constant.For example, when default
Audio format is WAV format, and when the audio format of voice messaging to be processed is also WAV format, then it does not need to carry out audio lattice
Formula conversion.
Format conversion unit 403c, if the audio format for the voice messaging to be processed is not preset audio format,
The audio format of the voice messaging to be processed is converted into preset audio format according to preset audio format transformation rule.
It wherein, is that basis is needed to preset when the audio format of voice messaging to be processed is not preset audio format
Audio format transformation rule audio format conversion is carried out to it.For example, the preset audio format transformation rule can be
Audio format conversion is carried out to voice messaging to be processed by Ffmpeg script.When preset audio format is WAV format, and wait locate
When the audio format for managing voice messaging is also MP3 format, the audio format by the voice to be processed is needed to convert.
Feature extraction unit 404 extracts each voice letter to be processed for calling preset voice screening script respectively
The characteristic information of breath, and the characteristic information of each voice messaging to be processed is respectively written into different readable text files.
Information judging unit 405, all readable text files for being successively read in first export folders
In characteristic information to judge whether the characteristic information in the readable text file matches with preset threshold.
First storage unit 406, if the characteristic information in the readable text file matches with preset threshold,
Voice messaging to be measured corresponding to the readable text file is stored into second export folders to be used for batch signatures.
As further embodiment, the message processing directives include preset third export folders address, described
Device 400 may further include the following units:
Type judging unit 407, all readable text files for being successively read in first export folders
In characteristic information with judge respectively the characteristic information in each readable text file type whether with preset characteristic information
Type match.
Wherein, according to the type of the type of the characteristic information of voice messaging to be processed and preset characteristic information whether
Match, so that it may judge whether it is effective voice messaging, if such as characteristic information need include voice messaging audio duration,
Sampling number, zoom factor and maximum amplitude value, and the preservation in readable text file only only has audio duration and adopts
Number of samples, then then the corresponding voice messaging to be processed of the readable text part is invalid information.And in readable text file
Save include audio duration, sampling number, zoom factor and maximum amplitude value, then show the readable text part it is corresponding to
Processing voice messaging is effective information.
Third judging unit 408, if type and preset feature for the characteristic information in the readable text file
The type of information do not match that, determines that voice messaging to be measured is invalid voice information corresponding to the readable text file, and
The voice messaging information to be measured is stored into third export folders corresponding to third export folders address.
Wherein, in order to further distinguish the property of voice messaging to be measured, it can will be determined as the to be measured of invalid voice information
Voice messaging is stored into third export folders corresponding to third export folders address.
As further embodiment, the message processing directives include the 4th export folders address, described device 400
Can also include:
Second storage unit 409, if the characteristic information in the readable text file is not matched that with preset threshold,
Then voice messaging to be measured corresponding to the readable text file is stored to corresponding to the 4th export folders address
In four export folders.
It should be noted that it is apparent to those skilled in the art that, the batch sieve of above-mentioned voice messaging
The specific implementation process of screening device 100 and each unit, can be with reference to the corresponding description in preceding method embodiment, for description
Convenienct and succinct, details are not described herein.
As seen from the above, in hardware realization, the above address acquisition unit 101, information determination unit 102, information are read
Unit 103, feature extraction unit 104, information judging unit 105 and first storage unit 106 etc. can be interior in the form of hardware
Be embedded in or the device reported a case to the security authorities independently of life insurance in, depositing for the batch screening plant of voice messaging can also be stored in a software form
In reservoir, the corresponding operation of above each unit is executed so that processor calls.The processor can be central processing unit
(CPU), microprocessor, single-chip microcontroller etc..
The batch screening plant of above-mentioned voice messaging can be implemented as a kind of form of computer program, and computer program can
To be run in computer equipment as shown in Figure 9.
Fig. 9 is a kind of structure composition schematic diagram of computer equipment of the present invention.The equipment can be server, wherein clothes
Business device can be independent server, be also possible to the server cluster of multiple server compositions.
Referring to Fig. 9, which includes processor 502, memory, the memory connected by system bus 501
Reservoir 504 and network interface 505, wherein memory may include non-volatile memory medium 503 and built-in storage 504.
The non-volatile memory medium 503 can storage program area 5031 and computer program 5032, the computer program
5032 are performed, and processor 502 may make to execute a kind of batch screening technique of voice messaging.
The processor 502 supports the operation of entire computer equipment 500 for providing calculating and control ability.
The built-in storage 504 provides environment for the operation of the computer program 5032 in non-volatile memory medium 503, should
When computer program 5032 is executed by processor 502, processor 502 may make to execute a kind of batch screening side of voice messaging
Method.
The network interface 505 is used to carry out network communication with other equipment.It will be understood by those skilled in the art that in Fig. 9
The structure shown, only the block diagram of part-structure relevant to application scheme, does not constitute and is applied to application scheme
The restriction of computer equipment 500 thereon, specific computer equipment 500 may include more more or fewer than as shown in the figure
Component perhaps combines certain components or with different component layouts.
Wherein, the processor 502 is for running computer program 5032 stored in memory, to realize institute as above
Step in the batch screening technique for the voice messaging stated.
It should be appreciated that in the embodiment of the present application, processor 502 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..Wherein, general processor can be microprocessor or
Person's processor is also possible to any conventional processor etc..
Those of ordinary skill in the art will appreciate that be realize above-described embodiment method in all or part of the process,
It is that relevant hardware can be instructed to complete by computer program.The computer program can be stored in a storage medium,
The storage medium is computer readable storage medium.The computer program is held by least one processor in the computer system
Row, to realize the process step of the embodiment of the above method.
Therefore, the present invention also provides a kind of storage mediums.The storage medium can be computer readable storage medium.This is deposited
Storage media is stored with computer program, which makes processor execute voice letter as described above when being executed by processor
Step in the batch screening technique of breath.
The storage medium can be USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), magnetic disk
Or the various computer readable storage mediums that can store program code such as CD.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided by the present invention, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary.For example, the division of each unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation.Such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.
The steps in the embodiment of the present invention can be sequentially adjusted, merged and deleted according to actual needs.This hair
Unit in bright embodiment device can be combined, divided and deleted according to actual needs.In addition, in each implementation of the present invention
Each functional unit in example can integrate in one processing unit, is also possible to each unit and physically exists alone, can also be with
It is that two or more units are integrated in one unit.
If the integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product,
It can store in one storage medium.Based on this understanding, technical solution of the present invention is substantially in other words to existing skill
The all or part of part or the technical solution that art contributes can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, terminal or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of batch screening technique of voice messaging, which is characterized in that the described method includes:
If receiving message processing directives, the address of the file where preset training set is obtained, and using the address as defeated
Enter folder address, the training set includes multiple voice messagings to be processed；
Preset threshold and preset first export folders address, the second output file are determined according to the message processing directives
Press from both sides address, wherein first export folders address is the address that the first export folders is saved, the first output text
Part folder includes multiple readable text files, and second export folders address is the address that the second export folders is saved；
The import folders address is read to obtain all voice messagings to be processed；
Call preset voice screening script to extract the characteristic information of each voice messaging to be processed respectively, and will be each wait locate
The characteristic information of reason voice messaging is respectively written into different readable text files；
Be successively read the characteristic information in all readable text files in first export folders with judge it is described can
Whether the characteristic information read in text file matches with preset threshold；
It, will be corresponding to the readable text file if the characteristic information in the readable text file matches with preset threshold
Voice messaging to be measured is stored into second export folders for batch signatures.
2. the method as described in claim 1, which is characterized in that it is described according to the message processing directives determine preset threshold with
And the step of preset first export folders address, the second export folders address, comprising:
The message processing directives are parsed to obtain corresponding presupposed information；
Preset threshold and preset first export folders address, the second export folders are determined according to the presupposed information
Location.
3. the method as described in claim 1, which is characterized in that the preset threshold includes preset duration threshold values, the feature
Information includes audio duration, the feature in all readable text files being successively read in first export folders
Information is the step of whether characteristic information in the readable text file matches with preset threshold judged, comprising:
The audio duration being successively read in all readable text files in first export folders；
Judge whether the audio duration in the readable text file is greater than or equal to preset duration threshold values；
If the audio duration in the readable text file is greater than or equal to preset duration threshold values, determine the machine readable text herein
Characteristic information in part matches with preset threshold.
4. method as claimed in claim 3, which is characterized in that the characteristic information determined in the readable text file with
Before the step of preset threshold matches, comprising:
If the audio duration in the readable text file is greater than or equal to preset duration threshold values, it is successively read first output
The sampling number in all readable text files in file；
Judge that the sampling number in the readable text file is greater than or equal to default sampling number；
If the sampling number in the readable text file is greater than or equal to default sampling number, determine the machine readable text herein
Characteristic information in part matches with preset threshold.
5. the method as described in claim 1, which is characterized in that described to call preset voice screening script every to extract respectively
The characteristic information of a voice messaging to be processed, and the characteristic information of each voice messaging to be processed is respectively written into different readable
Before step in text file, comprising:
It is every to determine respectively to be successively read the characteristic information in all readable text files in first export folders
The audio format of a voice messaging to be processed；
If the audio format of the voice messaging to be processed is preset audio format, the audio of the voice messaging to be processed is kept
Format is constant；
If the audio format of the voice messaging to be processed is not preset audio format, according to preset audio format transformation rule
The audio format of the voice messaging to be processed is converted into preset audio format.
6. method as claimed in claim 5, which is characterized in that the message processing directives include preset third output file
Address is pressed from both sides, the method also includes:
It is every to judge respectively to be successively read the characteristic information in all readable text files in first export folders
Whether the type of the characteristic information in a readable text file matches with the type of preset characteristic information；
If the type of the characteristic information in the readable text file and the type of preset characteristic information do not match that, institute is determined
Stating voice messaging to be measured corresponding to readable text file is invalid voice information, and by the voice messaging information to be measured store to
In third export folders corresponding to third export folders address.
7. the method as described in claim 1, which is characterized in that the message processing directives are including the 4th export folders
Location, the method also includes:
It, will be corresponding to the readable text file if the characteristic information in the readable text file is not matched that with preset threshold
Voice messaging to be measured store into the 4th export folders corresponding to the 4th export folders address.
8. a kind of batch screening plant of voice messaging, which is characterized in that including for executing such as any one of claim 1-7 institute
State the unit of method.
9. a kind of computer equipment, which is characterized in that the computer equipment includes memory and processor, on the memory
It is stored with computer program, the processor is realized as described in any one of claim 1-7 when executing the computer program
Method.
10. a kind of computer readable storage medium, which is characterized in that the storage medium is stored with computer program, the meter
Calculation machine program makes the processor execute such as method of any of claims 1-7 when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910197526.6A CN110059059B (en) | 2019-03-15 | 2019-03-15 | Batch screening method and device for voice information, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910197526.6A CN110059059B (en) | 2019-03-15 | 2019-03-15 | Batch screening method and device for voice information, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110059059A true CN110059059A (en) | 2019-07-26 |
| CN110059059B CN110059059B (en) | 2024-04-16 |
Family
ID=67316992
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910197526.6A Active CN110059059B (en) | 2019-03-15 | 2019-03-15 | Batch screening method and device for voice information, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110059059B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096941A (en) * | 2015-09-02 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Voice recognition method and device |
| WO2017008239A1 (en) * | 2015-07-14 | 2017-01-19 | 张阳 | Call control method and system for ktv song request system |
| CN107240395A (en) * | 2017-06-16 | 2017-10-10 | 百度在线网络技术（北京）有限公司 | A kind of acoustic training model method and apparatus, computer equipment, storage medium |
| WO2018107810A1 (en) * | 2016-12-15 | 2018-06-21 | 平安科技（深圳）有限公司 | Voiceprint recognition method and apparatus, and electronic device and medium |
-
2019
- 2019-03-15 CN CN201910197526.6A patent/CN110059059B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017008239A1 (en) * | 2015-07-14 | 2017-01-19 | 张阳 | Call control method and system for ktv song request system |
| CN105096941A (en) * | 2015-09-02 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Voice recognition method and device |
| WO2018107810A1 (en) * | 2016-12-15 | 2018-06-21 | 平安科技（深圳）有限公司 | Voiceprint recognition method and apparatus, and electronic device and medium |
| CN107240395A (en) * | 2017-06-16 | 2017-10-10 | 百度在线网络技术（北京）有限公司 | A kind of acoustic training model method and apparatus, computer equipment, storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110059059B (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6219643B1 (en) | Method of analyzing dialogs in a natural language speech recognition system | |
| CN106960051B (en) | Audio playing method and device based on electronic book and terminal equipment | |
| CN109360550A (en) | Test method, device, equipment and storage medium for voice interactive system | |
| CN109388675A (en) | Data analysing method, device, computer equipment and storage medium | |
| CN103136471A (en) | Method and system for testing malicious Android application programs | |
| WO2004029773A2 (en) | Software for statistical analysis of speech | |
| CN110209643A (en) | A kind of data processing method and device | |
| CN110047472A (en) | Batch conversion method, apparatus, computer equipment and the storage medium of voice messaging | |
| CN107492153B (en) | Attendance system, method, attendance server and attendance terminal | |
| CN111724908A (en) | Epidemic situation investigation method and device based on robot process automation RPA | |
| CN105895102A (en) | Recording editing method and recording device | |
| CN110246496A (en) | Speech recognition method, system, computer device and storage medium | |
| CN109523236A (en) | Mail generation method, device, computer equipment and storage medium | |
| CN110096479A (en) | Batch renaming method, apparatus, computer equipment and the storage medium of voice messaging | |
| CN110059139A (en) | Business datum archiving method, equipment, server and computer readable storage medium | |
| US20100172479A1 (en) | Dynamically improving performance of an interactive voice response (ivr) system using a complex events processor (cep) | |
| CN109389972A (en) | Quality detecting method, device, storage medium and the equipment of semantic cloud function | |
| CN110059059A (en) | Batch screening technique, device, computer equipment and the storage medium of voice messaging | |
| KR102195925B1 (en) | Method and apparatus for collecting voice data | |
| CN109101484A (en) | Recording file processing method, device, computer equipment and storage medium | |
| CN114697127B (en) | Service session risk processing method based on cloud computing and server | |
| CN111027319A (en) | Method and device for analyzing natural language time words and computer equipment | |
| CN110060667A (en) | Batch processing method, device, computer equipment and the storage medium of voice messaging | |
| CN111274156B (en) | Automatic identification method and device compatible with multi-frame pages | |
| CN109597948A (en) | Access method, system and the storage medium of URL link |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |