CN111079093A - Music score processing method and device and electronic equipment - Google Patents
Music score processing method and device and electronic equipment Download PDFInfo
- Publication number
- CN111079093A CN111079093A CN201911272496.7A CN201911272496A CN111079093A CN 111079093 A CN111079093 A CN 111079093A CN 201911272496 A CN201911272496 A CN 201911272496A CN 111079093 A CN111079093 A CN 111079093A
- Authority
- CN
- China
- Prior art keywords
- music score
- score
- vector
- processed
- music
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 29
- 239000013598 vector Substances 0.000 claims abstract description 176
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000012545 processing Methods 0.000 claims abstract description 15
- 230000005484 gravity Effects 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 8
- 238000011478 gradient descent method Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 11
- 238000012423 maintenance Methods 0.000 description 8
- 238000001914 filtration Methods 0.000 description 7
- 238000011176 pooling Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 241001342895 Chorus Species 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- HAORKNGNJCEJBX-UHFFFAOYSA-N cyprodinil Chemical compound N=1C(C)=CC(C2CC2)=NC=1NC1=CC=CC=C1 HAORKNGNJCEJBX-UHFFFAOYSA-N 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/10—Protecting distributed programs or content, e.g. vending or licensing of copyrighted material ; Digital rights management [DRM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Technology Law (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Auxiliary Devices For Music (AREA)
Abstract
The invention provides a music score processing method, a music score processing device and electronic equipment, wherein the method comprises the following steps: obtaining a music score to be processed; extracting a score vector of the music score to be processed through a convolutional neural network trained in advance; and recording the music score vector in a preset block chain system. In the method, a score vector of a score to be processed is extracted through a convolutional neural network trained in advance, and the score vector is recorded in a preset block chain system. The music score of each song is stored in a block chain system in a music score vector mode, and is permanently stored and can not be changed, so that the purpose of protecting the copyright of the music score is achieved.
Description
Technical Field
The present invention relates to the field of neural network technologies, and in particular, to a music score processing method and apparatus, and an electronic device.
Background
Music copyright maintenance has been a current social hotspot. In the related art, because the problem of plagiarism of music is difficult to define, the right of the music is difficult to be preserved, and the copyright of the music is difficult to be protected.
Disclosure of Invention
In view of the above, the present invention provides a music score processing method, apparatus and electronic device to protect the copyright of a music score.
In a first aspect, an embodiment of the present invention provides a music score processing method, including: obtaining a music score to be processed; extracting a score vector of the music score to be processed through a convolutional neural network trained in advance; and recording the music score vector in a preset block chain system.
In a preferred embodiment of the present invention, the step of extracting the score vector of the score to be processed through the pre-trained convolutional neural network includes: if the representation form of the music score to be processed is a staff sequence, converting the representation form of the music score to be processed into a digital sequence; and inputting the music score to be processed in a digital sequence form into a convolutional neural network trained in advance, and outputting the music score vector of the music score to be processed.
In a preferred embodiment of the present invention, the convolutional neural network includes a hidden layer; the hidden layer is constructed based on a gradient descent method; the convolutional neural network comprises a plurality of filter windows; the step of inputting the score to be processed in the form of a digital sequence into a convolutional neural network trained in advance and outputting the score vector of the score to be processed includes: inputting the music score to be processed in a digital sequence form into a hidden layer, and outputting a feature vector corresponding to the music score to be processed; inputting the characteristic vector into a convolutional neural network, and outputting a result vector corresponding to the characteristic vector by the convolutional neural network; the number of result vectors is the same as the number of filter windows; and calculating the specific gravity of each result vector based on the softmax function, and determining a score vector corresponding to the score to be processed based on the specific gravity.
In a preferred embodiment of the present invention, the step of determining the score vector corresponding to the score to be processed based on the specific gravity includes: sorting the result vectors according to the sequence of the specific gravity from large to small to obtain a sorting result; selecting a specified number of result vectors starting from a first result vector of the sorted results; and splicing the selected result vectors to obtain the music score vector corresponding to the music score to be processed.
In a preferred embodiment of the present invention, before the step of saving the score vector, the method further comprises: calculating the similarity between the music score vector and the stored music score vector based on a vector cosine similarity algorithm; if the similarity is larger than a preset similarity threshold, not storing the score vector; and if the similarity is not greater than the similarity threshold, saving the score vector.
In a preferred embodiment of the present invention, the method further includes: and sending the similarity corresponding to the score vector and the stored score vector.
In a preferred embodiment of the present invention, the method further includes: and sending the address information of the music score vector in the block chain system.
In a second aspect, an embodiment of the present invention further provides a musical score processing apparatus, including: the digital music score acquisition module is used for acquiring a music score to be processed; the music score vector output module is used for extracting the music score vector of the music score to be processed through a convolutional neural network which is trained in advance; and the music score vector storage module is used for recording the music score vector in a preset block chain system.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a processor and a memory, where the memory stores computer-executable instructions capable of being executed by the processor, and the processor executes the computer-executable instructions to implement the steps of the music score processing method described above.
In a fourth aspect, the embodiments of the present invention further provide a computer-readable storage medium, which stores computer-executable instructions, and when the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the steps of the music score processing method described above.
The embodiment of the invention has the following beneficial effects:
according to the music score processing method, the music score processing device and the electronic equipment, the music score vector of the music score to be processed is extracted through the convolutional neural network which is trained in advance, and the music score vector is recorded in the preset block chain system. The music score of each song is stored in a block chain system in a music score vector mode, and is permanently stored and can not be changed, so that the purpose of protecting the copyright of the music score is achieved.
Additional features and advantages of the disclosure will be set forth in the description which follows, or in part may be learned by the practice of the above-described techniques of the disclosure, or may be learned by practice of the disclosure.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a music score processing method according to an embodiment of the present invention;
fig. 2 is a flowchart of another music score processing method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a music score pre-training method according to an embodiment of the present invention;
FIG. 4 is a diagram of a score convolutional neural network according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a music score registration process according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a music score similarity query process according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a music score processing apparatus according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
At present, the problem of music score plagiarism of music is difficult to define, so that the right maintenance of the music score is difficult to carry out, and the copyright of the music score is difficult to protect. Based on this, the music score processing method, the music score processing device and the electronic device provided by the embodiment of the invention are applied to the technical field of the neural network and the block chain, and particularly relate to a method for learning a music score by using a convolutional neural network, which can obtain a music score vector, upload the music score vector to the block chain, and permanently store the music score vector, so that the purpose of protecting the copyright of the music score is realized.
For the convenience of understanding the embodiment, a music score processing method disclosed in the embodiment of the present invention will be described in detail first.
Example 1
An embodiment of the present invention provides a music score processing method, referring to a flowchart of the music score processing method shown in fig. 1, where the music score processing method includes the following steps:
and step S102, obtaining the music score to be processed.
The music score to be processed refers to the music score needing to be saved. A music score, i.e., a musical score, is a regular combination of various written symbols recording the pitch or tempo of music for editing and analyzing various melody formats. In the music field, whether domestic or foreign, music scores of various music works are often referred to as plagiarism, but the cases of really claiming to the court are rare.
The score generally has two representation forms of a staff and a number sequence, and the score to be processed in the embodiment generally refers to a score represented in a number sequence, and may also be referred to as a digital score or a digital numbered musical notation. The numbered musical notation is based on the mobile naming method, 7 basic levels in the scale are represented by 1, 2, 3, 4, 5, 6 and 7, the reading is do, re, mi, fa, sol, la and ti (or si), the English is represented by C, D, E, F, G, A, B, and the rest is represented by 0. The duration name of each digit corresponds to the 4-th note of the staff.
And step S104, extracting the score vector of the score to be processed through the convolutional neural network trained in advance.
A Convolutional Neural Network (CNN) is a feed-forward Neural Network whose artificial neurons can respond to a portion of the coverage of surrounding cells, and performs well for large image processing. The convolutional neural network consists of one or more convolutional layers and a top fully connected layer (corresponding to the classical neural network), and also includes associated weights and pooling layers (pooling layers). This structure enables the convolutional neural network to utilize a two-dimensional structure of the input data. Convolutional neural networks can give better results in terms of image and speech recognition than other deep learning structures.
And extracting a score vector of the score to be processed through a convolutional neural network, wherein the score vector is used for identifying the key and the design of the core of the score, and the recording of the score is equivalent to the recording of the key and the design of the core of the score.
Step S106, recording the music score vector in a preset block chain system.
The block chain is a concatenated text record (also called a block) cryptographically concatenated and protected with content. Each block contains the cryptographic hash, corresponding time stamp, and transaction data of the previous block, typically represented by a hash value computed using the Merkle tree algorithm, which is designed to render the contents of the block tamper-resistant. The distributed account book concatenated by the block chain technology can effectively record the transaction by two parties and permanently check the transaction. The music score vector is recorded by adopting a block chain system, so that the music score vector can be permanently stored and cannot be changed, and the purpose of protecting the copyright of the music score is achieved.
According to the music score processing method provided by the embodiment of the invention, the music score vector of the music score to be processed is extracted through the convolutional neural network which is trained in advance, and the music score vector is recorded in a preset block chain system. The music score of each song is stored in a block chain system in a music score vector mode, and is permanently stored and can not be changed, so that the purpose of protecting the copyright of the music score is achieved.
Example 2
The embodiment of the invention also provides another music score processing method; the method is realized on the basis of the method of the embodiment; the method mainly describes a specific implementation mode of extracting a score vector of a score to be processed through a convolutional neural network trained in advance.
Another score processing method as shown in fig. 2 is a flowchart of the score processing method, and the score processing method includes the steps of:
step S202, a music score to be processed is obtained.
And step S204, if the representation form of the music score to be processed is a staff series, converting the representation form of the music score to be processed into a digital series.
The representation form of the music score to be processed generally comprises a staff series and a number series, and if the representation form of the music score to be processed is the number series, the music score to be processed in the number series form is directly reserved without changing the representation form of the music score to be processed. If the representation form of the music score to be processed is a staff series, the music score to be processed in the staff form cannot be directly input into the convolutional neural network, and the music score to be processed in the staff form must be converted into the music score to be processed in the form of a digital series.
And step S206, inputting the music score to be processed in the form of a digital sequence into a convolutional neural network which is trained in advance, and outputting the music score vector of the music score to be processed.
The score to be processed input into the convolutional neural network and the score vector of the score to be processed can be output through the steps a 1-A3, which must be in the form of a numerical sequence:
step A1, inputting the score to be processed in the form of digit sequence into the hidden layer, and outputting the characteristic vector corresponding to the score to be processed.
Referring to a schematic diagram of a music score pre-training method shown in fig. 3, music score pre-training is implemented in the hidden layer, and before training a music score, the music score convolutional neural network first needs to vectorize the music score.
The embodiment divides the music score into 5 parts (respectively, a prelude of a song, a master song of a song, a chorus of a song, an interlude of songs and a tail of a song) in advance to obtain characteristic vectors of 5 aspects related to the music score. The input form of the score is a sequence of numbers of the score. The division of the score into 5 parts is generally performed in the form of labels, i.e. labels are added to the score, and the score is divided into 5 parts, for example: the 1-4 measure is marked as the prelude of the song, the 5-22 measure is the main song of the song, the 23-28 measure is the refrain of the song, the 29-32 measure is the interlude of the song, and the 33-40 measure is the tail of the song.
The invention uses a neural network with a hidden layer to directly train a music score vector to obtain a characteristic vector W5xdThen, the training is given to a convolution neural network. The hidden layer is constructed based on a gradient descent method, the pre-training algorithm uses the gradient descent method, data of the output layer is substituted into the weight of the input layer, W is continuously updated, and finally the characteristic vector W output by the hidden layer is a five-dimensional vector which corresponds to a song prelude, a song master song, a song aileron, a song interlude and a song tail.
Step A2, inputting the feature vector into a convolutional neural network, and outputting a result vector corresponding to the feature vector by the convolutional neural network; the number of result vectors is the same as the number of filter windows.
Referring to fig. 4, a schematic diagram of a score convolutional neural network is shown, which uses a plurality of filtering windows once W weights are trained. The first layer of the convolutional neural network is typically a convolutional layer. A filtering window (typically a window of size 3 x 3 or 5 x 5) is typically used to slide across the image, and this window area is the receptive field of the convolutional neural network. The filter window typically corresponds to a set of numbers, i.e. weights or parameters, wherein the depth of the filter window corresponds to the depth of the input image. The results of the filtering window correspond to the activation map or feature map.
The number and characteristics of the filtering windows used in the present embodiment are different according to the precision requirement. If the precision requirement is high, more filtering windows are used, meanwhile, the dimensionality of W is higher, and the dimensionality of the generated result vector is more. As shown in fig. 4, the example uses 4 filtering windows, obtains 4 result vectors, simplifies the vectors through the pooling layer and concatenates into one vector, and finally obtains two elements with the highest specific gravity through calculation by a softmax function as the vector of the score. A song may be adapted to produce different scores, but the core tune and design is still consistent. The invention learns the core of the music score through the convolutional neural network.
Pooling is another important concept in convolutional neural networks, which is actually a form of downsampling. There are many different forms of non-linear pooling functions, with "maximal pooling" being the most common.
And A3, calculating the specific gravity of each result vector based on the softmax function, and determining a score vector corresponding to the score to be processed based on the specific gravity.
The softmax function is used in a multi-classification process, which maps outputs of a plurality of neurons into a (0,1) interval, which can be understood as a probability, so as to perform multi-classification, as shown in fig. 4, a feature sequence W of a high input in fig. 4 is 5 rows and 6 columns (5 × 6), passes through 4 filtering windows, and corresponding convolutional layers, pooling layers, and flow exposure, respectively, so as to obtain 4 result vectors, the four feature vectors determine their respective specific gravities through the softmax function, and a score vector corresponding to a score to be processed is determined based on the specific gravities, which can be specifically performed through steps B1-B3:
b1, sorting the result vectors according to the sequence of the specific gravity from big to small to obtain a sorting result;
and sequencing the result vectors according to the sequence of the specific gravity from large to small to obtain a group of sequencing results, determining the specified number of the music score vectors extracted at this time by colleagues, wherein the specified number is generally determined by a user, and the specified number in the graph 4 is 2.
Step B2, starting with the first result vector of the sorted results, selecting a specified number of result vectors;
the first result vector of the sorting result is the result vector with the highest weight, a specified number of result vectors are selected according to the sorting result sequence, and corresponding to fig. 4, the first two result vectors of the sorting result are selected, and the score vector is constructed based on the two result vectors.
And step B3, splicing the selected result vectors to obtain a score vector corresponding to the score to be processed.
And splicing the selected result vectors to obtain a music score vector. This score vector can represent the melody and design of the score.
After determining the score vector of the score to be processed, it may be detected if the score is likely to be similar to the already saved score vector, performed by step C1-step C3:
and step C1, calculating the similarity between the score vector and the stored score vector based on the vector cosine similarity algorithm.
Step C2, if the similarity is larger than the preset similarity threshold, not saving the score vector;
and step C3, if the similarity is not larger than the similarity threshold, saving the score vector.
When a user registers an original music score (namely, a music score to be processed) to enter a background, the background firstly learns the music score vector according to the music score, then compares the music score vector with a background music database (used for storing the stored music score vector) according to the vector, if no similar music score is found (namely, the similarity is greater than a preset similarity threshold, and the similarity threshold is preferably 0.9), the user can register the music score to enter a block chain system to make permanent marking and recording for the copyright of the music score.
When the user checks whether a certain score is similar to his own score. The user uploads a target music score to a background, the background obtains a vector of the target music score through a learning system, then the target vector is compared with all music score vectors of the user, a comparison algorithm is a vector cosine similarity algorithm, and finally a similarity score list (namely, the similarity corresponding to the music score vector and the stored music score vector is sent) is obtained and returned to the user.
Step S208, record the score vector in a preset block chain system.
And if the music score is not similar to the stored music score, recording the music score vector corresponding to the music score in a preset block chain system. After saving, the address information of the music score in the block chain system can be sent to the owner of the music score.
Referring to a schematic diagram of a music book registration process shown in fig. 5, a user sends a music book to a digital asset management maintenance platform, a music book learning system in the digital asset management maintenance platform extracts the music book vector of the music book and ranks the music book vector into a block chain system, the block chain system includes a server cluster and a distributed account book, the block chain system returns an index to the digital asset management maintenance platform, and the index is address information of the music book vector in the block chain system. The digital asset management maintenance platform constructs an asset list based on the index, wherein the asset list comprises the index, the type (the type is the type of the digital asset, and in this embodiment, the musical score) and information (the information is basic information of the musical score uploaded by the user, such as the author of the musical score). After the asset list is generated, the digital asset management maintenance platform sends information that the registration was successful and the asset list to the user.
Referring to fig. 6, a diagram of a music score similarity query process is shown, and as shown in fig. 6, after a user uploads a new music score, the digital asset management maintenance platform may compare the similarity between a music score vector of the music score and a stored music score vector, and return a similarity result table to the user.
Example 3
Corresponding to the above method embodiment, an embodiment of the present invention provides a musical score processing apparatus, such as the structural schematic diagram of a musical score processing apparatus shown in fig. 7, the musical score processing apparatus including:
a digital music score obtaining module 71, configured to obtain a music score to be processed;
a score vector output module 72, configured to extract a score vector of a score to be processed through a pre-trained convolutional neural network;
the score vector storage module 73 records the score vector in a predetermined block chain system.
The music score processing device provided by the embodiment of the invention extracts the music score vector of the music score to be processed through the convolutional neural network trained in advance, and records the music score vector in the preset block chain system. The music score of each song is stored in a block chain system in a music score vector mode, and is permanently stored and can not be changed, so that the purpose of protecting the copyright of the music score is achieved.
In some embodiments, the score vector output module is configured to convert the representation form of the score to be processed into a number sequence if the representation form of the score to be processed is a staff series; and inputting the music score to be processed in a digital sequence form into a convolutional neural network trained in advance, and outputting the music score vector of the music score to be processed.
In some embodiments, the convolutional neural network comprises a hidden layer; the hidden layer is constructed based on a gradient descent method; the convolutional neural network comprises a plurality of filter windows; the music score vector output module is used for inputting the music score to be processed in a digital sequence form to the hidden layer and outputting a feature vector corresponding to the music score to be processed; inputting the characteristic vector into a convolutional neural network, and outputting a result vector corresponding to the characteristic vector by the convolutional neural network; the number of result vectors is the same as the number of filter windows; and calculating the specific gravity of each result vector based on the softmax function, and determining a score vector corresponding to the score to be processed based on the specific gravity.
In some embodiments, the score vector output module is configured to sort the result vectors in an order from a large specific gravity to a small specific gravity to obtain a sorted result; selecting a specified number of result vectors starting from a first result vector of the sorted results; and splicing the selected result vectors to obtain the music score vector corresponding to the music score to be processed.
In some embodiments, the apparatus further comprises a similarity comparison module for calculating similarity between the score vector and the stored score vector based on a vector cosine similarity algorithm; if the similarity is larger than a preset similarity threshold, not storing the score vector; and if the similarity is not greater than the similarity threshold, saving the score vector.
In some embodiments, the apparatus further includes a similarity sending module, configured to send a similarity between the score vector and the already stored score vector.
In some embodiments, the apparatus further includes an address information sending module, configured to send address information of the score vector in the block chain system.
The musical score processing apparatus provided by the embodiment of the present invention has the same technical features as the musical score processing method provided by the above embodiment, so that the same technical problems can be solved, and the same technical effects can be achieved.
Example 4
The embodiment of the invention also provides electronic equipment for operating the music score processing method; referring to fig. 8, an electronic device is shown, which includes a memory 100 and a processor 101, wherein the memory 100 is used for storing one or more computer instructions, and the one or more computer instructions are executed by the processor 101 to implement the music score processing method.
Further, the electronic device shown in fig. 8 further includes a bus 102 and a communication interface 103, and the processor 101, the communication interface 103, and the memory 100 are connected through the bus 102.
The Memory 100 may include a high-speed Random Access Memory (RAM) and may further include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The communication connection between the network element of the system and at least one other network element is realized through at least one communication interface 103 (which may be wired or wireless), and the internet, a wide area network, a local network, a metropolitan area network, and the like can be used. The bus 102 may be an ISA bus, PCI bus, EISA bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one double-headed arrow is shown in FIG. 8, but that does not indicate only one bus or one type of bus.
The processor 101 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 101. The Processor 101 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory 100, and the processor 101 reads the information in the memory 100, and completes the steps of the method of the foregoing embodiment in combination with the hardware thereof.
The embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the above-mentioned music score processing method.
The music score processing method, the music score processing apparatus, and the computer program product of the electronic device provided in the embodiments of the present invention include a computer-readable storage medium storing a program code, where instructions included in the program code may be used to execute the methods in the foregoing method embodiments, and specific implementation may refer to the method embodiments, and will not be described herein again.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the apparatus and/or the electronic device described above may refer to corresponding processes in the foregoing method embodiments, and are not described herein again.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A musical score processing method, comprising:
obtaining a music score to be processed;
extracting a score vector of the score to be processed through a pre-trained convolutional neural network;
and recording the music score vector in a preset block chain system.
2. The method of claim 1, wherein the step of extracting the score vector of the score to be processed through the pre-trained convolutional neural network comprises:
if the representation form of the music score to be processed is a staff series, converting the representation form of the music score to be processed into a digital series;
and inputting the music score to be processed in a digital sequence form into a convolutional neural network which is trained in advance, and outputting the music score vector of the music score to be processed.
3. The method of claim 2, wherein the convolutional neural network comprises a hidden layer; the hidden layer is constructed based on a gradient descent method; the convolutional neural network comprises a plurality of filter windows;
inputting the music score to be processed in a digital sequence form into a convolutional neural network which is trained in advance, and outputting a music score vector of the music score to be processed, wherein the step comprises the following steps:
inputting the music score to be processed in a digital sequence form into the hidden layer, and outputting a feature vector corresponding to the music score to be processed;
inputting the feature vector to the convolutional neural network, and outputting a result vector corresponding to the feature vector by the convolutional neural network; the number of result vectors is the same as the number of filter windows;
and calculating the specific gravity of each result vector based on a softmax function, and determining a score vector corresponding to the score to be processed based on the specific gravity.
4. The method of claim 3, wherein the step of determining the score vector corresponding to the to-be-processed score based on the specific gravity comprises:
sorting the result vectors according to the sequence of the specific gravity from large to small to obtain a sorting result;
selecting a specified number of said result vectors starting with a first of said result vectors of said ordered results;
and splicing the selected result vectors to obtain the music score vector corresponding to the music score to be processed.
5. The method of claim 4, wherein prior to the step of saving the score vector, the method further comprises:
calculating the similarity between the music score vector and the saved music score vector based on a vector cosine similarity algorithm;
if the similarity is larger than a preset similarity threshold value, the score vector is not stored;
and if the similarity is not greater than the similarity threshold value, saving the score vector.
6. The method of claim 5, further comprising:
and sending the similarity corresponding to the music score vector and the saved music score vector.
7. The method of claim 1, further comprising:
and sending the address information of the music score vector in a block chain system.
8. A musical score processing apparatus, comprising:
the digital music score acquisition module is used for acquiring a music score to be processed;
the music score vector output module is used for extracting the music score vector of the music score to be processed through a convolutional neural network which is trained in advance;
and the music score vector storage module is used for recording the music score vector in a preset block chain system.
9. An electronic device, comprising a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the steps of the music score processing method of any of claims 1 to 7.
10. A computer-readable storage medium storing computer-executable instructions which, when invoked and executed by a processor, cause the processor to carry out the steps of the score processing method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911272496.7A CN111079093B (en) | 2019-12-11 | 2019-12-11 | Music score processing method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911272496.7A CN111079093B (en) | 2019-12-11 | 2019-12-11 | Music score processing method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111079093A true CN111079093A (en) | 2020-04-28 |
| CN111079093B CN111079093B (en) | 2022-07-01 |
Family
ID=70314099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911272496.7A Active CN111079093B (en) | 2019-12-11 | 2019-12-11 | Music score processing method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111079093B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101657816A (en) * | 2007-02-14 | 2010-02-24 | 缪斯亚米有限公司 | The portal website that is used for distributed audio file editing |
| CN106446952A (en) * | 2016-09-28 | 2017-02-22 | 北京邮电大学 | Method and apparatus for recognizing score image |
| US9918141B2 (en) * | 2015-08-05 | 2018-03-13 | Surewaves Mediatech Private Limited | System and method for monitoring and detecting television ads in real-time using content databases (ADEX reporter) |
| CN108681602A (en) * | 2018-05-22 | 2018-10-19 | 王翔 | A kind of spectrum power supply management method |
| CN110264987A (en) * | 2019-06-18 | 2019-09-20 | 王子豪 | Chord based on deep learning carries out generation method |
-
2019
- 2019-12-11 CN CN201911272496.7A patent/CN111079093B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101657816A (en) * | 2007-02-14 | 2010-02-24 | 缪斯亚米有限公司 | The portal website that is used for distributed audio file editing |
| US9918141B2 (en) * | 2015-08-05 | 2018-03-13 | Surewaves Mediatech Private Limited | System and method for monitoring and detecting television ads in real-time using content databases (ADEX reporter) |
| CN106446952A (en) * | 2016-09-28 | 2017-02-22 | 北京邮电大学 | Method and apparatus for recognizing score image |
| CN108681602A (en) * | 2018-05-22 | 2018-10-19 | 王翔 | A kind of spectrum power supply management method |
| CN110264987A (en) * | 2019-06-18 | 2019-09-20 | 王子豪 | Chord based on deep learning carries out generation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111079093B (en) | 2022-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109885692B (en) | Knowledge data storage method, apparatus, computer device and storage medium | |
| CN109978060B (en) | Training method and device of natural language element extraction model | |
| Huang et al. | Object-location-aware hashing for multi-label image retrieval via automatic mask learning | |
| CN108228704A (en) | Identify method and device, the equipment of Risk Content | |
| CN110750640A (en) | Text data classification method and device based on neural network model and storage medium | |
| CN112766288B (en) | Image processing model construction method, device, electronic equipment and readable storage medium | |
| CN112819073B (en) | Classification network training, image classification method and device and electronic equipment | |
| CN112149386A (en) | Event extraction method, storage medium and server | |
| CN113806548A (en) | Petition factor extraction method and system based on deep learning model | |
| JP2023501010A (en) | A Classification Method for Application Preference Text Based on TextRank | |
| CN113515620A (en) | Method and device for sorting technical standard documents of power equipment, electronic equipment and medium | |
| CN117453917A (en) | Model training method and device, storage medium and electronic equipment | |
| CN109902162B (en) | Text similarity identification method based on digital fingerprints, storage medium and device | |
| CN113032524A (en) | Trademark infringement identification method, terminal device and storage medium | |
| CN111079093B (en) | Music score processing method and device and electronic equipment | |
| CN112507120B (en) | Prediction method and device for keeping classification consistency | |
| CN112559679B (en) | Political new media propagation force detection method, device, equipment and storage medium | |
| CN114398980A (en) | Cross-modal Hash model training method, encoding method, device and electronic equipment | |
| JP5824429B2 (en) | Spam account score calculation apparatus, spam account score calculation method, and program | |
| CN114817586A (en) | Target object classification method and device, electronic equipment and storage medium | |
| CN114417860A (en) | Information detection method, device and equipment | |
| CN113868417A (en) | Sensitive comment identification method and device, terminal equipment and storage medium | |
| CN112801221B (en) | Data classification method, device, equipment and storage medium | |
| CN111178817A (en) | Judgment result obtaining method and device based on deep learning | |
| CN111680513B (en) | Feature information identification method and device and computer readable storage medium |
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 |