CN108053815A - The performance control method and robot of a kind of robot - Google Patents
The performance control method and robot of a kind of robot Download PDFInfo
- Publication number
- CN108053815A CN108053815A CN201711322453.6A CN201711322453A CN108053815A CN 108053815 A CN108053815 A CN 108053815A CN 201711322453 A CN201711322453 A CN 201711322453A CN 108053815 A CN108053815 A CN 108053815A
- Authority
- CN
- China
- Prior art keywords
- robot
- melody
- target
- unit
- performance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10H—ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
- G10H1/00—Details of electrophonic musical instruments
- G10H1/0033—Recording/reproducing or transmission of music for electrophonic musical instruments
- G10H1/0041—Recording/reproducing or transmission of music for electrophonic musical instruments in coded form
- G10H1/0058—Transmission between separate instruments or between individual components of a musical system
- G10H1/0066—Transmission between separate instruments or between individual components of a musical system using a MIDI interface
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Acoustics & Sound (AREA)
- Manipulator (AREA)
- Toys (AREA)
Abstract
The present invention relates to field of artificial intelligence, and in particular to the performance control method and robot of a kind of robot, including:Robot, which receives, plays instruction, performance instruction is including at least the melody title of target melody, it obtains and parses the original MIDI files to match with the melody title, obtain the performance control parameter of target melody, and the built-in camera captured image of control robot, and the environment being presently according to pattern recognition machine device people, then control parameter will be played to be adjusted to play control parameter with the target that the environment that robot is presently in matches, control parameter control robot is played according to target and plays target melody.As it can be seen that the embodiment of the present invention, can make robot under different performance occasions using different playing modes, so as to improve the user experience of listener.
Description
Technical field
The present invention relates to field of artificial intelligence, and in particular to the performance control method and robot of a kind of robot.
Background technology
Development in science and technology with rapid changepl. never-ending changes and improvements is filled with many fresh elements to music field, gradually change people write, table
It drills, make the mode for even appreciating music, wherein, have benefited from the rapid development of artificial intelligence technology, robot can also be as drilling
As playing house, it is people's playing music, brings different audio experience.For robot, machine can be passed through
The method of device study learns the version that different concert performers plays same melody, is derived from playing the playing mode of the melody,
Playing mode includes playing performance speed, playing strength of the melody etc..It however, can be in different performances compared to concert performer is played
Occasion plays same a piece of music using different playing modes, and robot can only be drilled when playing with a piece of music by fixed
The mode of playing is played, and the playing mode of robot is excessively single.
The content of the invention
The embodiment of the present invention discloses the performance control method and robot of a kind of robot, can make robot different
It plays using different playing modes under occasion, so as to improve the user experience of listener.
The robot, which receives, plays instruction, described to play melody title of the instruction including at least target melody;
The robot obtains the original MIDI files to match with the melody title;
The robot parses the original MIDI files, obtains the performance control parameter of the target melody;
The robot controls the built-in camera captured image of the robot, and identifies the machine according to described image
The environment that device people is presently in;
The performance control parameter is adjusted to what is matched with the environment that the robot is presently in by the robot
Target plays control parameter, and plays control parameter according to the target and the robot is controlled to play the target melody.
As an alternative embodiment, in first aspect of the embodiment of the present invention, the robot is according to the mesh
After mark plays the control parameter control robot performance target melody, the method further includes:
Whether new performance instruction, the starting of the preset time period are received in the robot detection preset time period
At the time of moment is that the target music playing terminates;
It waits to drill at least one of the target melody similarity higher than predetermined threshold value if it is not, then the robot is obtained
It plays music song;
The robot is output by voice the melody title of at least one melody to be performed, so that user is according to extremely
The melody title selection target melody to be performed of a few first melody to be performed.
As an alternative embodiment, in first aspect of the embodiment of the present invention, the method further includes:
The robot builds convolutional neural networks model, and the convolutional neural networks model is used to identify the robot
Residing environment;
The robot obtains Massive Sample image, and forms training sample set according to the Massive Sample image;
Robot convolutional neural networks model according to the training sample set pair is trained, and is trained
Convolutional neural networks model;
The environment that the robot is presently according to the described image identification robot, including:
The robot is identified described image using the trained convolutional neural networks model, to identify
The environment that the robot is presently in.
As an alternative embodiment, in first aspect of the embodiment of the present invention, the method further includes:
During the target melody is played, the robot control sound pick-up is recorded described in the robot performance
Audio signal during target melody;
The robot determines noise signal from the audio signal, obtains the characteristic parameter of the noise signal,
And noise reduction process is carried out to the audio signal according to the characteristic parameter, obtain audio file;
The robot detects whether there is the mobile terminal that communication connection is established with the robot;
If it is present the robot sends sharing information to the mobile terminal, the sharing information includes described
Audio file.
As an alternative embodiment, in first aspect of the embodiment of the present invention, the method further includes:
The corresponding position coordinates of environment that the robot acquisition is presently in;
The audio file and the position coordinates are associated by the robot, obtain the audio file with it is described
The incidence relation of position coordinates;
The robot shares the incidence relation to internet terminal, so that the internet terminal is detecting
It states at position coordinates there are during the mobile terminal of other users, recommends the audio text to the mobile terminal of other users
Part.
Second aspect of the embodiment of the present invention discloses a kind of robot, and the robot includes:
Receiving unit plays instruction for receiving, described to play melody title of the instruction including at least target melody;
First acquisition unit, for obtaining the original MIDI files to match with the melody title;
Resolution unit for parsing the original MIDI files, obtains the performance control parameter of the target melody;
Recognition unit for controlling the built-in camera captured image of the robot, and identifies institute according to described image
State the environment that robot is presently in;
Adjustment unit, for the performance control parameter to be adjusted to match with the environment that the robot is presently in
Target play control parameter;
Unit is played, for playing control parameter according to the target robot is controlled to play the target melody.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the robot further includes:
First detection unit controls the robot for playing control parameter according to the target in the performance unit
After playing the target melody, detect and new performance instruction whether is received in preset time period, the preset time period
At the time of initial time is that the target music playing terminates;
Second acquisition unit, for detecting to be not received by institute in the preset time period in the first detection unit
When stating new performance instruction, at least one melody to be performed for being higher than predetermined threshold value with the target melody similarity is obtained;
Output unit, for being output by voice the melody title of at least one melody to be performed, so that user's root
According to the melody title selection target melody to be performed of melody to be performed at least one described.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the robot further includes:
Construction unit, for building convolutional neural networks model, the convolutional neural networks model is used to identify the machine
Environment residing for device people;
Unit is formed, for obtaining Massive Sample image, and training sample set is formed according to the Massive Sample image;
Training unit is trained for the convolutional neural networks model according to the training sample set pair, is instructed
The convolutional neural networks model perfected;
The recognition unit, specifically for the trained convolutional neural networks mould obtained using the training unit
Described image is identified in type, to identify environment that the robot is presently in.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the robot further includes:
Recording unit, during playing the target melody in the performance unit, control sound pick-up record institute
State the audio signal when target melody is played by robot;
Processing unit for determining noise signal from the audio signal, obtains the feature ginseng of the noise signal
Number, and noise reduction process is carried out to the audio signal according to the characteristic parameter, obtain audio file;
Second detection unit, for detecting whether in the presence of the mobile terminal that communication connection is established with the robot;
Transmitting element, for detecting to exist the shifting that communication connection is established with the robot in the second detection unit
During dynamic terminal, sharing information is sent to the mobile terminal, the sharing information includes the audio file.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the first acquisition unit is also used
In the corresponding position coordinates of environment that acquisition is presently in;
The robot further includes:
Associative cell, for the audio file and the position coordinates to be associated, obtain the audio file with
The incidence relation of the position coordinates;
Shares unit, for sharing the incidence relation to internet terminal, so that the internet terminal is detecting
To at the position coordinates there are during the mobile terminal of other users, recommend the audio to the mobile terminal of other users
File.
The third aspect of the embodiment of the present invention discloses another robot, including:
It is stored with the memory of executable program code;
The processor coupled with the memory;
The processor calls the executable program code stored in the memory, performs the embodiment of the present invention the
A kind of performance control method of robot disclosed in one side.
Fourth aspect of the embodiment of the present invention discloses a kind of computer readable storage medium, stores computer program, wherein,
The computer program causes computer to perform a kind of performance controlling party of robot disclosed in first aspect of the embodiment of the present invention
Method.
The 5th aspect of the embodiment of the present invention discloses a kind of computer program product, when the computer program product is calculating
When being run on machine so that the computer performs the part or all of step of any one method of first aspect.
The aspect of the embodiment of the present invention the 6th disclose a kind of using distribution platform, and the application distribution platform is for issuing calculating
Machine program product, wherein, when the computer program product is run on computers so that the computer performs first party
The part or all of step of any one method in face.
Compared with prior art, the embodiment of the present invention has the advantages that:
In the embodiment of the present invention, robot, which receives, plays instruction, and performance instruction is including at least the melody name of target melody
Claim, obtain and parse the original MIDI files to match with the melody title, obtain target melody performance control parameter and
The built-in camera captured image of robot, and the environment being presently according to pattern recognition machine device people are controlled, then will be played
Control parameter is adjusted to play control parameter with the target that the environment that robot is presently in matches, and is played and controlled according to target
Target melody is played by state modulator robot.As it can be seen that implementing the embodiment of the present invention, can be drilled according to the different adjustment of ambient enviroment
Control parameter is played, so that robot uses different playing mode playing musics under different performance occasions, so as to improve
The user experience of listener.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment
Attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some embodiments of the present invention, for ability
For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is a kind of flow diagram of the performance control method of robot disclosed by the embodiments of the present invention;
Fig. 2 is the flow diagram of the performance control method of another robot disclosed by the embodiments of the present invention;
Fig. 3 is the flow diagram of the performance control method of another robot disclosed by the embodiments of the present invention;
Fig. 4 is a kind of structure diagram of robot disclosed by the embodiments of the present invention;
Fig. 5 is the structure diagram of another robot disclosed by the embodiments of the present invention;
Fig. 6 is the structure diagram of another robot disclosed by the embodiments of the present invention;
Fig. 7 is the structure diagram of another robot disclosed by the embodiments of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, the every other reality that those of ordinary skill in the art are obtained without making creative work
Example is applied, belongs to the scope of protection of the invention.
It should be noted that term " comprising " and " having " and their any changes in the embodiment of the present invention and attached drawing
Shape, it is intended that cover non-exclusive include.Such as contain the process of series of steps or unit, method, system, product or
The step of equipment is not limited to list or unit, but optionally further include the step of not listing or unit or optional
Ground is further included for the intrinsic other steps of these processes, method, product or equipment or unit.
The embodiment of the present invention discloses the performance control method and robot of a kind of robot, can make robot different
It plays using different playing modes under occasion, so as to improve the user experience of listener.It is described in detail individually below.
Embodiment one
Referring to Fig. 1, Fig. 1 is a kind of flow signal of the performance control method of robot disclosed by the embodiments of the present invention
Figure.As shown in Figure 1, the performance control method of the robot can include following operation:
101st, robot, which receives, plays instruction, and performance instruction is including at least the melody title of target melody.
In the embodiment of the present invention, performance instruction can be that user is sent out by the mobile terminal bound in advance with the robot
It send, the form of performance instruction can be that text can also be voice, and the embodiment of the present invention does not limit.For example, use
Family can be parent, and the mobile terminal bound in advance with the robot can be the mobile phone of parent, then, when parent goes out and child
When son is in alone, parent wishes that robot can play piano pieces for child, moulds sentiment, the culture child couple of child with this
The interest of music, so parent can send voice messaging by mobile phone to robot, which wishes machine including parent
The melody title for the target melody that device people plays.After robot receives the voice messaging of parent's transmission, to the voice messaging
It is identified, and the corresponding piano music of melody title is played according to the entitled child of melody of the target melody identified.It can
See, the embodiment of the present invention, can realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance appreciate, improve user experience.
102nd, robot obtains the original MIDI files to match with above-mentioned melody title.
In the embodiment of the present invention, MIDI (Musical Instrument Digital Interface), also referred to as musical instrument
Digital interface is the electronic communication protocol of an industrial standard, is that the definition of the performance apparatus (such as synthesizer) such as electronic musical instrument is various
Note plays code, and electronic musical instrument, computer, mobile phone or the outfit of other stage performance is allowed to be connected to each other, adjustment with it is synchronous, obtain
To exchange such performance data immediately.For MIDI files, MIDI files are broadly divided into as File header information (Header
Chunk) and two parts of track information (Track Chunk), the basic letter in File header information major storage MIDI files
Breath, such as:Each crotchet is cut into how many points (Tick), track information block counts, file format code names etc., and sound
Rail information is then used for storing a series of MIDI information, and according to different file format code names, MIDI information can be used for representing
Single track (file format code name 0) or the MIDI information of multiple tracks (file format code name 1), the embodiment of the present invention is not
It limits.
As an alternative embodiment, before the original MIDI files to match with above-mentioned melody title are obtained,
The audio file that robot can be matched by handling above-mentioned melody title, generation match original with above-mentioned melody title
MIDI files, specifically, robot can pair audio file to match with above-mentioned melody title handle, obtain the sound
The time attribute description information of frequency file, the time attribute description information are the corresponding time categories of each note in audio file
Property;Then sub-frame processing is carried out to the audio file according to predetermined time period, obtains each audio frame in the audio file
Barycenter of frequency spectrum, and the spectrum mean of audio file is calculated according to the barycenter of frequency spectrum and place is normalized to the spectrum mean
Reason, obtains MIDI file parameters, which can be used for the accuracy in pitch information for describing the audio file;Robot can be according to this
The corresponding time attribute of each note and the generation of the accuracy in pitch information of the audio file and above-mentioned melody title in audio file
The original MIDI files to match.As it can be seen that the embodiment of the present invention, can be calculated by audio file and generated and the audio file
Corresponding MIDI files, improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process.
103rd, robot parses original MIDI files, obtains the performance control parameter of above-mentioned target melody.
In the embodiment of the present invention, robot parses original MIDI files, can obtain MIDI file parameters, wherein, it should
MIDI file parameters can include above-mentioned target melody all notes and all notes in each note risen with target melody
The note datas such as interval time (offset), pitch (frequency), duration (duration) and dynamics at the beginning, the MIDI files
Parameter is the performance control parameter of above-mentioned target melody.
In the embodiment of the present invention, optionally, robot parses original MIDI files, can also obtain above-mentioned target melody
Whole tone color.Robot can adjust the whole tone color of the target melody according to the arbitrary tone color of user's selection, then exchange
Whole tone color after whole carries out the adaptation processing of corresponding parameter.
104th, robot controls the built-in camera captured image of the robot, and current according to pattern recognition machine device people
Residing environment.
In the embodiment of the present invention, robot can control built-in camera captured image, it is preferred that robot can control
Built-in camera photographing panorama picture, i.e. robot can preset coverage and set multiple panorama focusing positions, and lead to
Built-in camera shooting multiple focus images corresponding with above-mentioned multiple panorama focusing positions are crossed, and by this multiple focus images life
Into panoramic picture and the corresponding panoramic picture of default coverage.
105th, above-mentioned performance control parameter is adjusted to the mesh to match with the environment that the robot is presently in by robot
Mark plays control parameter, and plays control parameter according to target and the robot is controlled to play above-mentioned target melody.
In the embodiment of the present invention, control parameter is played according to target, the robot is controlled to play above-mentioned target melody, specifically
For, when target melody is played by robot, target can be played control parameter and be converted into pulse output signals, that is, read in target
The note data in control parameter is played, the interval of each note and target melody section start in the note data, that is, target melody
Time (offset), pitch (frequency), duration (duration) and dynamics etc., and each note is formed according to the note data
Corresponding pulse signal, and the pulse signal of different notes is corresponded to different signal output channels, and signal output channels
Incidence relation is established with the motor of robot.For example, when robot is by piano performance melody, robot can be according to
The time sequencing output pulse signal of each note, can be started with driving motor, i.e. robot strikes up target melody.
As it can be seen that by the described methods of Fig. 1, control ginseng can be played according to the adjustment of the different intelligent of ambient enviroment
Number, so that robot uses different playing mode playing musics under different performance occasions, so as to improve listener's
User experience;
Furthermore, it is possible to realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance is appreciated, and improves user experience;And it can be calculated and generated corresponding with the audio file by audio file
MIDI files improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process.
Embodiment two
Referring to Fig. 2, Fig. 2 is the flow signal of the performance control method of another robot disclosed by the embodiments of the present invention
Figure.As shown in Fig. 2, the performance control method of the robot may comprise steps of:
In the embodiment of the present invention, the performance control method of the robot further includes step 201~205, for step 201~
205 description, refer to the detailed description that step 101~105 are directed in embodiment one, and the embodiment of the present invention repeats no more.
206th, new performance instruction whether is received in robot detection preset time period, if it is, being drilled according to new
It plays instruction and plays a first new melody, whereas if it is no, then perform step 207.
In the embodiment of the present invention, at the time of the initial time of above-mentioned preset time period is that target music playing terminates.
As an alternative embodiment, robot is preset with user information database, prestored in the user's information bank to
The legal facial image and/or legal voiceprint data of a few validated user and each validated user like melody type
Data, wherein, the legal facial image and/or legal voiceprint data of each validated user are with liking melody similar material mutually to tie up
It is fixed.For example, the environment that robot is presently in is family, and new drill is not received when robot is detected in preset time period
When playing instruction, robot can start face identification functions and/or Application on Voiceprint Recognition function, obtain the photo and/or vocal print of user,
And by the legal facial image of the photo of the user and/or vocal print with all validated users in user information database and/or legal sound
Texel data is matched, if the photo and/or vocal print of the user in user information database any one validated user it is legal
Facial image and/or legal voiceprint data match, then, robot can be with the conjunction of the successful match in calling and obtaining user information bank
Method user's likes melody similar material, and analyzes this and like melody similar material, is pushed to user and likes melody type with this
The melody that data matches.Particularly, when robot start it is that face identification functions and/or Application on Voiceprint Recognition identification of function go out, simultaneously
And when being more than one with the number of users of validated user successful match in user information database, robot can be by calculating these use
The registration of melody similar material is liked to select push melody in family.As it can be seen that the embodiment of the present invention, can pass through face
And/or Application on Voiceprint Recognition goes out after validated user, the music style data to be prestored according to validated user, provides personalization to the user
Melody is recommended, and improves the Experience Degree of user.
207th, robot obtains at least one melody to be performed for being higher than predetermined threshold value with above-mentioned target melody similarity.
208th, robot is output by voice the melody title of at least one melody to be performed, so that user is according at least one
The melody title selection target melody to be performed of first melody to be performed.
For step 206~step 208, robot can pre-set preset time period, the starting of the preset time period
Moment is that upper a piece of music plays the finish time terminated.So, after a piece of music performance finishes, if robot is pre-
If receiving new performance instruction in the period, robot can be drilled according to the new melody title that new performance instruction includes
Play the new melody;If robot does not receive new performance instruction in preset time period, robot can pass through analysis
Above-mentioned target melody is selected to user's push with the high melody to be performed of above-mentioned target melody similarity for user.As it can be seen that this
Inventive embodiments can not select next melody to be performed in user, it is intelligent to user's push and user it
The melody that the target melody music style of preceding selection is similar is selected for user, you can with according to the push of the user preferences intelligence use
The interested melody in family, while improve user experience.
As it can be seen that by the described methods of Fig. 2, control ginseng can be played according to the adjustment of the different intelligent of ambient enviroment
Number, so that robot uses different playing mode playing musics under different performance occasions, so as to improve listener's
User experience;
Furthermore, it is possible to realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance is appreciated, and improves user experience;And it can be calculated and generated corresponding with the audio file by audio file
MIDI files improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process;
Furthermore, it is possible to after validated user is gone out by face and/or Application on Voiceprint Recognition, the melody that is prestored according to validated user
Style data provides personalized melody to the user and recommends, improves the Experience Degree of user;
Furthermore, it is possible to next melody to be performed is not selected in user, it is intelligent to user's push and user
The similar melody of the target melody music style that selects before is selected for user, you can with according to the push of user preferences intelligence this
The interested melody of user, while improve user experience.
Embodiment three
Referring to Fig. 3, Fig. 3 is the flow signal of the performance control method of another robot disclosed by the embodiments of the present invention
Figure.As shown in figure 3, the performance control method of the robot may comprise steps of:
301st, robot structure convolutional neural networks model, the convolutional neural networks model are used to identify residing for robot
Environment.
In the embodiment of the present invention, robot structure convolutional neural networks model, specifically, robot can be designed by defeated
Enter the convolutional neural networks model of this five layers of neutral nets compositions of layer, three hidden layers and output layer, and each layer of neutral net
Corresponding different weight parameter value.For example, robot is when by the content of the convolutional neural networks Model Identification image,
The first layer of convolutional neural networks can find the edge in the image;The second layer can find the edge institute group that first layer is found
Into figure, such as rectangle or circle;Third layer, which can be found, has symbolic characteristic in special characteristic, such as image
Article;Image is transferred to next layer by each layer after the target of this layer is searched out, until last layer, and final output
As a result codetermined by weight parameter value all in neutral net.
302nd, robot obtains Massive Sample image, and forms training sample set according to Massive Sample image.
In the embodiment of the present invention, the sample that training sample is concentrated can be used for training neural network model.
303rd, robot is trained according to the above-mentioned convolutional neural networks model of training sample set pair, obtains trained volume
Product neural network model.
In the embodiment of the present invention, convolutional neural networks are a kind of feedforward neural networks, its artificial neuron can respond
Surrounding cells in a part of coverage are mainly used for identification, the processing aspect of image.
In the embodiment of the present invention, the performance control method of the robot further includes step 304~306, for step 304~
306 description, refer to the detailed description that step 101~103 are directed in embodiment one, and the embodiment of the present invention repeats no more.
307th, robot controls the built-in camera captured image of the robot, and uses above-mentioned trained convolutional Neural
The image is identified in network model, to identify environment that robot is presently in.
In the embodiment of the present invention, robot is identified the image using above-mentioned trained convolutional neural networks model
Concrete mode can be:Robot controls the built-in camera captured image of the robot, it is preferred that the built-in camera can
To shoot the 360 of robot ° of panoramic picture, then robot can pre-process the panoramic picture, i.e.,
Gray processing processing is carried out to the panoramic picture, gray level image is obtained after processing, the gray level image is then inputted into trained volume
Product neural network model makes the gray level image successively by input layer, the knowledge of three hidden layers and output layer this five layers of neutral nets
Not, by exporting recognition result after the convolutional neural networks Model Identification.For example, the environment that robot is presently in is to drill
Play the Room, then, robot can shoot the panoramic picture of ambient enviroment by built-in camera, and use trained convolution god
The panoramic picture is identified through network model, which identifies that the environment is to drill successively by layering
Play the Room, then output recognition result is that the environment that the robot is presently in is concert hall.As it can be seen that the embodiment of the present invention, it can
Image to be identified by trained convolutional neural networks model, to identify environment that robot is presently in, phase
Compared with traditional images identification method to the complicated characteristic extraction procedure of image, image recognition processes are simplified, so as to also improve
The efficiency and accuracy of environment-identification.
308th, above-mentioned performance control parameter is adjusted to the mesh to match with the environment that the robot is presently in by robot
Mark plays control parameter, and plays control parameter according to target and the robot is controlled to play above-mentioned target melody.
In the embodiment of the present invention, due to factors such as the spatial position environment, noise intensity, listenership of different environment not
Together, robot can set database, which prestores multiple performance control parameters for playing music, and plays
Control parameter is associated with environment scene, i.e., different environment scenes matches different performance control parameters.For example, ring
Border scene has concert hall, classroom and family etc., then, the performance to match with concert hall can be prestored in database and is controlled and is joined
Number and the classroom performance control parameter to match and the performance control parameter to match with family, when robot is according to step 308
When to identify the environment that robot is presently in be concert hall, the concert hall can be obtained from database match and play control
Parameter, and the performance control parameter is determined as target and plays control parameter, and control parameter is played according to target and controls the machine
Device people plays above-mentioned target melody.As it can be seen that the embodiment of the present invention, can be directed to the performance of different environmental suitability adjustment melodies
Control parameter so that robot obtains the different deduction modes to same a piece of music, improves the intelligent of playing music, makes pleasure
Bent performance is more bonded environment.
309th, during above-mentioned target melody is played, robot control sound pick-up records the robot and plays above-mentioned mesh
Mark audio signal during melody.
In the embodiment of the present invention, robot can control sound pick-up to record the machine during target melody is played
People plays audio signal during above-mentioned target melody, and the initial time that sound pick-up records the audio signal is that target is played by robot
The initial time of melody, the end time that sound pick-up records the audio signal are the end time that target melody is played by robot,
Wherein, sound pick-up is a device for being transmitted to rear end equipment again for collection site ambient sound, it is by microphone (Mike
Wind) and audio amplifier circuit composition.
310th, robot determines noise signal from above-mentioned audio signal, obtains the characteristic parameter of noise signal, and root
Noise reduction process is carried out to the audio signal according to characteristic parameter, obtains audio file.
In the embodiment of the present invention, get sound pick-up record audio signal after, robot to the audio signal into
Row Signal Pretreatment is exactly that robot can carry out spectrum analysis to the audio signal, to determine that the audio is believed specifically
Noise signal in number, and the characteristic parameter of noise signal is obtained, the audio signal is carried out at noise reduction according to this feature parameter
Reason, obtains audio file.Optionally, after robot determines the noise signal in audio signal, can also be believed according to noise
Number phase carries out reverse phase inhibition processing to noise signal, obtains audio file.As it can be seen that the embodiment of the present invention, can believe from audio
Noise signal is isolated in number, realizes the noise reduction to audio signal, so as to obtain the more pure audio file of tonequality.
311st, robot detects whether there is the mobile terminal that communication connection is established with the robot, if it is present holding
Row step 312, whereas if being not present, terminates this flow.
312nd, robot sends sharing information to mobile terminal, which includes above-mentioned audio file.
For step 309~step 312, for example, the environment that robot is presently in is advanced concert hall, mobile whole
End is the mobile phone of audience.After audience is in the target melody that advanced concert hall listening robot is played, to this target melody
Performance version it is very satisfied, it is desirable to be able to by the performance version be retained in the mobile phone of oneself so as to the spare time appreciate, that
, audience can pass through mobile phone before advanced concert hall is entered and robot is established and communicated to connect;Robot can play
Audio signal when playing the target melody is recorded while target melody and noise reduction process obtains audio file, and
The audio file is sent on the mobile phone of the audience.As it can be seen that the embodiment of the present invention, the audio that can will play target melody is literary
Part carries out being shared with the mobile terminal for establishing communication connection with robot after noise reduction process, allow the user of mobile terminal with
When listen to the audio file, realize the interaction between robot and the user of mobile terminal, improve the Experience Degree of user.
313rd, the corresponding position coordinates of environment that robot acquisition is presently in.
314th, above-mentioned audio file and above-mentioned position coordinates are associated by robot, obtain audio file and position coordinates
Incidence relation.
In the embodiment of the present invention, above-mentioned audio file and above-mentioned position coordinates are associated by robot, obtain audio text
The incidence relation of part and position coordinates establishes the correspondence between audio file and position coordinates.
315th, robot shares above-mentioned incidence relation to internet terminal, so that internet terminal is detecting position seat
There are during the mobile terminal of other users at mark, recommend above-mentioned audio file to the mobile terminal of other users.
For step 313~step 315, can be made by the incidence relation formed between audio file and position coordinates
It obtains once detecting at the position coordinates there are during the mobile terminal of other users, recommends above-mentioned sound to the mobile terminal immediately
Frequency file, so as to fulfill the shared of audio file.
Wherein, the performance control method of the robot further includes step 316~318, for the description of step 316~318,
The detailed description that step 206~208 are directed in embodiment one is refer to, the embodiment of the present invention repeats no more.
As it can be seen that by the described methods of Fig. 3, control ginseng can be played according to the adjustment of the different intelligent of ambient enviroment
Number, so that robot uses different playing mode playing musics under different performance occasions, so as to improve listener's
User experience;
Furthermore, it is possible to realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance is appreciated, and improves user experience;And it can be calculated and generated corresponding with the audio file by audio file
MIDI files improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process;
Furthermore, it is possible to after validated user is gone out by face and/or Application on Voiceprint Recognition, the melody that is prestored according to validated user
Style data provides personalized melody to the user and recommends, improves the Experience Degree of user;
Furthermore, it is possible to next melody to be performed is not selected in user, it is intelligent to user's push and user
The similar melody of the target melody music style that selects before is selected for user, you can with according to the push of user preferences intelligence this
The interested melody of user, while improve user experience;
Furthermore, it is possible to image is identified by trained convolutional neural networks model, to identify that robot works as
Preceding residing environment compared to complicated characteristic extraction procedure of the traditional images identification method to image is played, simplifies image identification
Process, so as to also improve the efficiency of environment-identification and accuracy;
Furthermore, it is possible to the performance control parameter for different environmental suitability adjustment melodies so that robot acquisition pair
With the different deduction modes of a piece of music, the intelligent of playing music is improved, the performance of melody is made more to be bonded environment;
Furthermore, it is possible to isolate noise signal from audio signal, the noise reduction to audio signal is realized, so as to obtain tonequality
More pure audio file;
Lead to furthermore, it is possible to which the audio file for playing target melody is carried out noise reduction process and is shared with to establish with robot afterwards
Believe the mobile terminal of connection, the user of mobile terminal is allow to listen to the audio file at any time, it is whole with movement to realize robot
Interaction between the user at end improves the Experience Degree of user;It and can be by being formed between audio file and position coordinates
Incidence relation so that it is whole to the movement immediately once detecting at the position coordinates there are during the mobile terminal of other users
Above-mentioned audio file is recommended at end, so as to fulfill the shared of audio file.
Example IV
Referring to Fig. 4, Fig. 4 is a kind of structure diagram of robot disclosed by the embodiments of the present invention.It as shown in figure 4, should
Robot can include:
Receiving unit 401 plays instruction for receiving, and performance instruction includes at least the melody title of target melody, and
The melody title is supplied to first acquisition unit 402.
In the embodiment of the present invention, it is advance with the robot that performance instruction that receiving unit 401 receives can be that user passes through
What the mobile terminal of binding was sent, the form of performance instruction can be that text can also be voice, and the embodiment of the present invention is not done
It limits.For example, user can be parent, and the mobile terminal bound in advance with the robot can be the mobile phone of parent, that
, when parent goes out and child is in alone, parent wishes that robot can play piano pieces for child, moulds child with this
Sentiment, culture child to the interest of music, so parent can send voice messaging, voice letter by mobile phone to robot
Breath includes the melody title that parent wishes the target melody that robot is played.Robot receive parent transmission voice messaging it
Afterwards, which is identified, and the melody title is played according to the entitled child of melody of the target melody identified
Corresponding piano music.As it can be seen that the embodiment of the present invention, can realize that remote control robot plays pleasure for child when parent is outgoing
Song brings real piano performance to appreciate, improves user experience for child.
First acquisition unit 402, for obtaining the original MIDI files to match with above-mentioned melody title.
As an alternative embodiment, first acquisition unit 402 is in the original obtained and above-mentioned melody title matches
Before beginning MIDI file, the audio file that can also be matched by handling above-mentioned melody title, generation and above-mentioned melody title
The original MIDI files to match, specifically, robot can pair audio file to match with above-mentioned melody title progress
Processing, obtains the time attribute description information of the audio file, which is each sound in audio file
Accord with corresponding time attribute;Then sub-frame processing is carried out to the audio file according to predetermined time period, obtains the audio file
In each audio frame barycenter of frequency spectrum, and the spectrum mean of audio file and equal to the frequency spectrum is calculated according to the barycenter of frequency spectrum
Value is normalized, and obtains MIDI file parameters, which can be used for the accuracy in pitch letter for describing the audio file
Breath;Robot can be according to the corresponding time attribute of each note in the audio file and the accuracy in pitch information of the audio file
The original MIDI files that generation matches with above-mentioned melody title.As it can be seen that the embodiment of the present invention, can be calculated by audio file
And MIDI files corresponding with the audio file are generated, the efficiency of audio frequency process is improved, and then also improves audio frequency process
It is intelligent.
Resolution unit 403 for parsing the original MIDI files that above-mentioned first acquisition unit 402 is got, obtains above-mentioned
The performance control parameter of target melody, and trigger recognition unit 404 and start.
In the embodiment of the present invention, resolution unit 403 parses original MIDI files, can obtain MIDI file parameters, wherein,
The MIDI file parameters can include above-mentioned target melody all notes and all notes in each note and target melody
The note datas such as interval time (offset), pitch (frequency), duration (duration) and the dynamics of section start, MIDI texts
Part parameter is the performance control parameter of above-mentioned target melody.
In the embodiment of the present invention, optionally, resolution unit 403 parses original MIDI files, can also obtain above-mentioned target
The whole tone color of melody.Robot can adjust the whole tone color of the target melody according to the arbitrary tone color of user's selection, so
The adaptation for carrying out corresponding parameter to the whole tone color after adjustment afterwards is handled.
Recognition unit 404, for controlling the built-in camera captured image of robot, and according to pattern recognition machine device people
The environment being presently in, and trigger and start adjustment unit 405.
Adjustment unit 405, the performance control parameter for above-mentioned resolution unit 403 to be obtained are adjusted to work as with the robot
The target that preceding residing environment matches plays control parameter.
Unit 406 is played, for playing control parameter according to the target of above-mentioned adjustment unit 405 robot is controlled to play
Target melody.
In the embodiment of the present invention, play unit 406 and according to target performance control parameter the robot is controlled to play above-mentioned mesh
Melody is marked, specifically, when playing the performance target melody of unit 406, it is defeated target performance control parameter can be converted into pulse
Go out signal, that is, read in target and play note data in control parameter, each note and mesh in the note data, that is, target melody
Interval time (offset), pitch (frequency), duration (duration) and dynamics of melody section start etc. is marked, and according to the sound
Symbol data form the corresponding pulse signal of each note, and the pulse signal of different notes is corresponded to different signal outputs and is led to
Road, and the motor of signal output channels and robot establishes incidence relation.For example, piano performance melody is passed through in robot
When, playing unit 406 can be started, i.e. machine according to the time sequencing output pulse signal of each note with driving motor
People strikes up target melody.
As it can be seen that by the described robots of Fig. 4, control ginseng can be played according to the adjustment of the different intelligent of ambient enviroment
Number, so that robot uses different playing mode playing musics under different performance occasions, so as to improve listener's
User experience;
Furthermore, it is possible to realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance is appreciated, and improves user experience;And it can be calculated and generated corresponding with the audio file by audio file
MIDI files improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process.
Embodiment five
Referring to Fig. 5, Fig. 5 is the structure diagram of another robot provided in an embodiment of the present invention, wherein, Fig. 5 institutes
The robot shown is that robot as shown in Figure 4 further optimizes.Compared with robot shown in Fig. 4, Fig. 5
Shown robot can also include:
First detection unit 407 is drilled for playing control parameter control robot according to target in above-mentioned performance unit 406
After playing target melody, detect and new performance instruction, the initial time of the preset time period whether are received in preset time period
At the time of being that target music playing terminates.
As an alternative embodiment, first detection unit 407 is preset with user information database, in the user's information bank
Prestore the legal facial image of at least one validated user and/or liking for legal voiceprint data and each validated user
Melody similar material, wherein, the legal facial image and/or legal voiceprint data of each validated user are with liking melody type to provide
Material is mutually bound.For example, the environment that robot is presently in is family, is not received when in robot detection preset time period
When new performance instructs, first detection unit 407 can start face identification functions and/or Application on Voiceprint Recognition function, obtain user
Photo and/or vocal print, and by the legal face of the photo of the user and/or vocal print and all validated users in user information database
Image and/or legal voiceprint data are matched, if the photo and/or vocal print of the user and any one position in user information database
The legal facial image and/or legal voiceprint data of validated user match, then, first detection unit 407 can transfer use
The validated user of the successful match likes melody similar material in the information bank of family, and triggers second acquisition unit 408 and start, by
Second acquisition unit 408 analyzes this and likes melody similar material, is pushed to user and likes what melody similar material matched with this
Melody.Particularly, when robot starts face identification functions and/or Application on Voiceprint Recognition identification of function go out and and user information
When the number of users of validated user successful match is more than one in storehouse, robot can like melody by calculate these users
The registration of similar material come select push melody.As it can be seen that the embodiment of the present invention, can pass through face and/or Application on Voiceprint Recognition
Go out after validated user, the music style data to be prestored according to validated user, provide personalized melody to the user and recommend, improve
The Experience Degree of user.
Second acquisition unit 408, for detecting to be not received by preset time period in above-mentioned first detection unit 407
When new performance instructs, at least one melody to be performed for being higher than predetermined threshold value with target melody similarity is obtained, and is triggered defeated
Go out unit 409 to start.
Output unit 409, for being output by voice the melody title of at least one melody to be performed so that user according to
The melody title selection target melody to be performed of at least one melody to be performed.
As it can be seen that by the described robots of Fig. 5, control ginseng can be played according to the adjustment of the different intelligent of ambient enviroment
Number, so that robot uses different playing mode playing musics under different performance occasions, so as to improve listener's
User experience;
Furthermore, it is possible to realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance is appreciated, and improves user experience;And it can be calculated and generated corresponding with the audio file by audio file
MIDI files improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process;
Furthermore, it is possible to after validated user is gone out by face and/or Application on Voiceprint Recognition, the melody that is prestored according to validated user
Style data provides personalized melody to the user and recommends, improves the Experience Degree of user;
Furthermore, it is possible to next melody to be performed is not selected in user, it is intelligent to user's push and user
The similar melody of the target melody music style that selects before is selected for user, you can with according to the push of user preferences intelligence this
The interested melody of user, while improve user experience.
Embodiment six
Referring to Fig. 6, Fig. 6 is the structure diagram of another robot disclosed by the embodiments of the present invention.Wherein, Fig. 6 institutes
The robot shown is that robot as shown in Figure 5 optimizes.Compared with the robot shown in Fig. 5, shown in Fig. 6
Robot further includes:
Construction unit 410, for building convolutional neural networks model, which is used to identify robot
Residing environment, and triggering forms unit 411 and starts.
In the embodiment of the present invention, construction unit 410 builds convolutional neural networks model, and specifically, robot can be set
The convolutional neural networks model that meter is made of input layer, three hidden layers and output layer this five layers of neutral nets, and each layer of god
Different weight parameter values is corresponded to through network.For example, robot is passing through the convolutional neural networks Model Identification image
During content, the first layer of convolutional neural networks can find the edge in the image;The second layer can find what first layer was found
The figure that edge is formed, such as rectangle or circle;Third layer, which can be found, has mark in special characteristic, such as image
The article of property feature;Image is transferred to next layer by each layer after the target of this layer is searched out, until last layer, and most
Whole output is codetermined by weight parameter value all in neutral net.
Unit 411 is formed, for obtaining Massive Sample image, and training sample set is formed according to Massive Sample image.
Training unit 412, for the above-mentioned construction unit 410 of training sample set pair formed according to above-mentioned formation unit 411
The convolutional neural networks model of structure is trained, and obtains trained convolutional neural networks model.
Above-mentioned recognition unit 404, specifically for using above-mentioned 412 trained convolutional neural networks model pair of training unit
Image is identified, to identify environment that the robot is presently in.
In the embodiment of the present invention, recognition unit 404 uses above-mentioned 412 trained convolutional neural networks model of training unit
Can be to the concrete mode that the image is identified:Control the built-in camera captured image of the robot, it is preferred that this is interior
360 ° of panoramic picture of robot can be shot by putting camera, then the panoramic picture be pre-processed, i.e.,
Gray processing processing is carried out to the panoramic picture, gray level image is obtained after processing, the gray level image is then inputted into trained volume
Product neural network model makes the gray level image successively by input layer, the knowledge of three hidden layers and output layer this five layers of neutral nets
Not, by exporting recognition result after the convolutional neural networks Model Identification.For example, the environment that robot is presently in is to drill
Play the Room, then, recognition unit 404 can shoot the panoramic picture of ambient enviroment by built-in camera, and pass through recognition unit
404 are identified the panoramic picture using trained convolutional neural networks model, and the convolutional neural networks model is by dividing
Layer identifies that the environment is concert hall successively, then it is that the robot is presently in that recognition unit 404, which exports recognition result,
Environment is concert hall.As it can be seen that the embodiment of the present invention, can know image by trained convolutional neural networks model
Not, to identify environment that robot is presently in, compared to playing complicated feature extraction of the traditional images identification method to image
Process simplifies image recognition processes, so as to also improve the efficiency of environment-identification and accuracy.
As an alternative embodiment, the robot shown in Fig. 6 can also include:
Recording unit 413, during playing target melody in above-mentioned performance unit 406, control sound pick-up record
Audio signal during target melody is played by robot, and the audio signal is supplied to processing unit 414.
Processing unit 414 for determining noise signal from above-mentioned audio signal, obtains the feature ginseng of the noise signal
Number, and noise reduction process is carried out to above-mentioned audio signal according to characteristic parameter, audio file is obtained, and triggers second detection unit
415 start.
In the embodiment of the present invention, after the audio signal that recording unit 413 records is obtained, processing unit 414 is to the sound
Frequency signal carries out Signal Pretreatment, is exactly that processing unit 414 can carry out spectrum analysis to the audio signal specifically, with
It determines the noise signal in the audio signal, and obtains the characteristic parameter of noise signal, according to this feature parameter to the audio
Signal carries out noise reduction process, obtains audio file.Optionally, noise signal of the processing unit 414 in audio signal is determined
Afterwards, reverse phase inhibition processing can also be carried out to noise signal, obtain audio file according to noise signal phase.As it can be seen that this hair
Bright embodiment can isolate noise signal from audio signal, realize the noise reduction to audio signal, so as to obtain tonequality more
Pure audio file.
Second detection unit 415, for detecting whether in the presence of the mobile terminal that communication connection is established with the robot.
Transmitting element 416 establishes communication connection for detecting to exist in above-mentioned second detection unit 415 with robot
During mobile terminal, sharing information is sent to mobile terminal, which includes the audio text that above-mentioned processing unit 414 obtains
Part.
Above-mentioned first acquisition unit 402 is additionally operable to obtain the corresponding position coordinates of environment being presently in.
Optionally, the robot shown in Fig. 6 further includes:
Associative cell 417, for audio file and the above-mentioned first acquisition unit 402 for obtaining above-mentioned processing unit 414
The position coordinates got is associated, and obtains the incidence relation of the audio file and the position coordinates, and by the incidence relation
It is supplied to shares unit 418.
Shares unit 418, for sharing incidence relation to internet terminal, so that internet terminal is detecting the position
It puts at coordinate there are during the mobile terminal of other users, recommends above-mentioned audio file to the mobile terminal of other users.
As it can be seen that by the described robots of Fig. 6, control ginseng can be played according to the adjustment of the different intelligent of ambient enviroment
Number, so that robot uses different playing mode playing musics under different performance occasions, so as to improve listener's
User experience;
Furthermore, it is possible to realize that remote control robot is child's playing music when parent is outgoing, brings really for child
Piano performance is appreciated, and improves user experience;And it can be calculated and generated corresponding with the audio file by audio file
MIDI files improve the efficiency of audio frequency process, and then also improve the intelligent of audio frequency process;
Furthermore, it is possible to after validated user is gone out by face and/or Application on Voiceprint Recognition, the melody that is prestored according to validated user
Style data provides personalized melody to the user and recommends, improves the Experience Degree of user;
Furthermore, it is possible to next melody to be performed is not selected in user, it is intelligent to user's push and user
The similar melody of the target melody music style that selects before is selected for user, you can with according to the push of user preferences intelligence this
The interested melody of user, while improve user experience;
Furthermore, it is possible to image is identified by trained convolutional neural networks model, to identify that robot works as
Preceding residing environment compared to complicated characteristic extraction procedure of the traditional images identification method to image is played, simplifies image identification
Process, so as to also improve the efficiency of environment-identification and accuracy;
Furthermore, it is possible to the performance control parameter for different environmental suitability adjustment melodies so that robot acquisition pair
With the different deduction modes of a piece of music, the intelligent of playing music is improved, the performance of melody is made more to be bonded environment;
Furthermore, it is possible to isolate noise signal from audio signal, the noise reduction to audio signal is realized, so as to obtain tonequality
More pure audio file;
Lead to furthermore, it is possible to which the audio file for playing target melody is carried out noise reduction process and is shared with to establish with robot afterwards
Believe the mobile terminal of connection, the user of mobile terminal is allow to listen to the audio file at any time, it is whole with movement to realize robot
Interaction between the user at end improves the Experience Degree of user;It and can be by being formed between audio file and position coordinates
Incidence relation so that it is whole to the movement immediately once detecting at the position coordinates there are during the mobile terminal of other users
Above-mentioned audio file is recommended at end, so as to fulfill the shared of audio file.
Referring to Fig. 7, Fig. 7 is the structure diagram of another robot disclosed by the embodiments of the present invention.As shown in fig. 7,
The robot can include:
It is stored with the memory 701 of executable program code;
The processor 702 coupled with memory 701;
Wherein, processor 702 calls the executable program code stored in memory 701, and it is any one to perform Fig. 1~Fig. 3
The performance control method of kind robot.
The embodiment of the present invention discloses a kind of computer readable storage medium, stores computer program, wherein, the computer
Program causes computer to perform the performance control method of any one robot of Fig. 1~Fig. 3.
It is to be understood that " one embodiment " or " embodiment " that specification is mentioned in the whole text mean it is related with embodiment
A particular feature, structure, or characteristic is included at least one embodiment of the present invention.Therefore, occur everywhere in entire disclosure
" in one embodiment " or " in one embodiment " identical embodiment is not necessarily referred to.In addition, these special characteristics, structure
Or characteristic can in any suitable manner combine in one or more embodiments.Those skilled in the art should also know,
Embodiment described in this description belongs to alternative embodiment, and not necessarily the present invention must for involved action and module
Must.
In various embodiments of the present invention, it should be appreciated that the size of the sequence number of above-mentioned each process is not meant to perform suitable
The inevitable priority of sequence, the execution sequence of each process should be determined with its function and internal logic, without tackling the embodiment of the present invention
Implementation process forms any restriction.
In various embodiments of the present invention, it should be appreciated that the terms "and/or" is only a kind of description association pair
The incidence relation of elephant, expression may have three kinds of relations, such as A and/or B, can represent:Individualism A, exist simultaneously A and
These three situations of B, individualism B.In addition, character "/" herein, it is a kind of pass of "or" to typically represent forward-backward correlation object
System.
In embodiment provided by the present invention, it should be appreciated that " B corresponding with A " represents that B is associated with A, can be with according to A
Determine B.It is also to be understood that determine that B is not meant to determine B only according to A according to A, it can also be according to A and/or other information
Determine B.
The above-mentioned unit illustrated as separating component may or may not be physically separate, be shown as unit
The component shown may or may not be physical location, you can positioned at a place or can also be distributed to multiple nets
On network unit.Part or all of unit therein can be selected to realize the purpose of this embodiment scheme according to the actual needs.
In addition, each functional unit in various embodiments of the present invention can be integrated in a processing unit or
Unit is individually physically present, can also two or more units integrate in a unit.Above-mentioned integrated unit
Both the form that hardware may be employed is realized, can also be realized in the form of SFU software functional unit.
If above-mentioned integrated unit realized in the form of SFU software functional unit and be independent production marketing or in use,
It can be stored in a retrievable memory of computer.Based on such understanding, technical scheme substantially or
Person says all or part of of the part contribute to the prior art or the technical solution, can be in the form of software product
It embodies, which is stored in a memory, including several requests with so that a computer is set
Standby (can be personal computer, server or network equipment etc., can be specifically the processor in computer equipment) performs
The part or all of step of each embodiment above method of the present invention.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
Relevant hardware to be instructed to complete by program, which can be stored in a computer readable storage medium, storage
Medium include read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory,
RAM), programmable read only memory (Programmable Read-only Memory, PROM), erasable programmable is read-only deposits
Reservoir (Erasable Programmable Read Only Memory, EPROM), disposable programmable read-only memory (One-
Time Programmable Read-Only Memory, OTPROM), the electronics formula of erasing can make carbon copies read-only memory
(Electrically-Erasable Programmable Read-Only Memory, EEPROM), read-only optical disc (Compact
Disc Read-Only Memory, CD-ROM) or other disk storages, magnetic disk storage, magnetic tape storage or can
For carrying or storing any other computer-readable medium of data.
Detailed Jie has been carried out to the performance control method and robot of a kind of robot disclosed by the embodiments of the present invention above
It continues, specific case used herein is set forth the principle of the present invention and embodiment, and the explanation of above example is only
It is the method and its core concept for being used to help understand the present invention;Meanwhile for those of ordinary skill in the art, according to this hair
Bright thought, there will be changes in specific embodiments and applications, in conclusion this specification content should not manage
It solves as limitation of the present invention.
Claims (10)
1. a kind of performance control method of robot, which is characterized in that the described method includes:
The robot, which receives, plays instruction, described to play melody title of the instruction including at least target melody;
The robot obtains the original MIDI files to match with the melody title;
The robot parses the original MIDI files, obtains the performance control parameter of the target melody;
The robot controls the built-in camera captured image of the robot, and identifies the robot according to described image
The environment being presently in;
The performance control parameter is adjusted to the target to match with the environment that the robot is presently in by the robot
Control parameter is played, and control parameter is played according to the target, the robot is controlled to play the target melody.
2. according to the method described in claim 1, it is characterized in that, control parameter control is played according to the target by the robot
After making the robot performance target melody, the method further includes:
Whether new performance instruction, the initial time of the preset time period are received in the robot detection preset time period
At the time of being that the target music playing terminates;
If it is not, then the robot obtains at least one pleasure to be performed for being higher than predetermined threshold value with the target melody similarity
It is bent;
The robot is output by voice the melody title of at least one melody to be performed, so that user is according at least one
The melody title selection target melody to be performed of the first melody to be performed.
3. method according to claim 1 or 2, which is characterized in that the method further includes:
The robot builds convolutional neural networks model, and the convolutional neural networks model is used to identify residing for the robot
Environment;
The robot obtains Massive Sample image, and forms training sample set according to the Massive Sample image;
Robot convolutional neural networks model according to the training sample set pair is trained, and obtains trained volume
Product neural network model;
The environment that the robot is presently according to the described image identification robot, including:
The robot is identified described image using the trained convolutional neural networks model, described to identify
The environment that robot is presently in.
4. according to claim 1-3 any one of them methods, which is characterized in that the method further includes:
During the target melody is played, the robot control sound pick-up records the robot and plays the target
Audio signal during melody;
The robot determines noise signal from the audio signal, obtains the characteristic parameter of the noise signal, and root
Noise reduction process is carried out to the audio signal according to the characteristic parameter, obtains audio file;
The robot detects whether there is the mobile terminal that communication connection is established with the robot;
If it is present the robot sends sharing information to the mobile terminal, the sharing information includes the audio
File.
5. according to the method described in claim 4, it is characterized in that, the method further includes:
The corresponding position coordinates of environment that the robot acquisition is presently in;
The audio file and the position coordinates are associated by the robot, obtain the audio file and the position
The incidence relation of coordinate;
The robot shares the incidence relation to internet terminal so that the internet terminal detecting rheme
It puts at coordinate there are during the mobile terminal of other users, recommends the audio file to the mobile terminal of other users.
6. a kind of robot, which is characterized in that including:
Receiving unit plays instruction for receiving, described to play melody title of the instruction including at least target melody;
First acquisition unit, for obtaining the original MIDI files to match with the melody title;
Resolution unit for parsing the original MIDI files, obtains the performance control parameter of the target melody;
Recognition unit for controlling the built-in camera captured image of the robot, and identifies the machine according to described image
The environment that device people is presently in;
Adjustment unit, for the performance control parameter to be adjusted to the mesh to match with the environment that the robot is presently in
Mark plays control parameter;
Unit is played, for playing control parameter according to the target robot is controlled to play the target melody.
7. robot according to claim 6, which is characterized in that the robot further includes:
First detection unit controls the robot to play for playing control parameter according to the target in the performance unit
After the target melody, detect and new performance instruction, the starting of the preset time period whether are received in preset time period
At the time of moment is that the target music playing terminates;
Second acquisition unit is described new for detecting to be not received by the preset time period in the first detection unit
Performance instruction when, obtain with the target melody similarity be higher than predetermined threshold value at least one melody to be performed;
Output unit, for being output by voice the melody title of at least one melody to be performed, so that user is according to extremely
The melody title selection target melody to be performed of a few first melody to be performed.
8. the robot according to claim 6 or 7, which is characterized in that the robot further includes:
Construction unit, for building convolutional neural networks model, the convolutional neural networks model is used to identify the robot
Residing environment;
Unit is formed, for obtaining Massive Sample image, and training sample set is formed according to the Massive Sample image;
Training unit is trained for the convolutional neural networks model according to the training sample set pair, is trained
Convolutional neural networks model;
The recognition unit, specifically for the trained convolutional neural networks model pair obtained using the training unit
Described image is identified, to identify environment that the robot is presently in.
9. according to claim 6-8 any one of them robot, which is characterized in that the robot further includes:
Recording unit, during playing the target melody in the performance unit, control sound pick-up records the machine
Device people plays the audio signal during target melody;
Processing unit for determining noise signal from the audio signal, obtains the characteristic parameter of the noise signal, and
Noise reduction process is carried out to the audio signal according to the characteristic parameter, obtains audio file;
Second detection unit, for detecting whether in the presence of the mobile terminal that communication connection is established with the robot;
Transmitting element, for detecting to exist the mobile end that communication connection is established with the robot in the second detection unit
During end, sharing information is sent to the mobile terminal, the sharing information includes the audio file.
10. robot according to claim 9, which is characterized in that the first acquisition unit is additionally operable to obtain current institute
The corresponding position coordinates of environment at place;
The robot further includes:
Associative cell, for the audio file and the position coordinates to be associated, obtain the audio file with it is described
The incidence relation of position coordinates;
Shares unit, for sharing the incidence relation to internet terminal, so that the internet terminal is detecting
It states at position coordinates there are during the mobile terminal of other users, recommends the audio text to the mobile terminal of other users
Part.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711322453.6A CN108053815A (en) | 2017-12-12 | 2017-12-12 | The performance control method and robot of a kind of robot |
PCT/CN2017/117681 WO2019114015A1 (en) | 2017-12-12 | 2017-12-21 | Robot performance control method and robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711322453.6A CN108053815A (en) | 2017-12-12 | 2017-12-12 | The performance control method and robot of a kind of robot |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108053815A true CN108053815A (en) | 2018-05-18 |
Family
ID=62131964
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711322453.6A Pending CN108053815A (en) | 2017-12-12 | 2017-12-12 | The performance control method and robot of a kind of robot |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108053815A (en) |
WO (1) | WO2019114015A1 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108922315A (en) * | 2018-08-21 | 2018-11-30 | 福州大学厦门工艺美术学院 | A kind of guitar auxiliary practice device |
CN110136677A (en) * | 2019-03-28 | 2019-08-16 | 深圳芒果未来教育科技有限公司 | Musical sound control method and Related product |
WO2021068812A1 (en) * | 2019-10-12 | 2021-04-15 | 平安科技(深圳)有限公司 | Music generation method and apparatus, electronic device and computer-readable storage medium |
WO2024008217A1 (en) * | 2023-06-08 | 2024-01-11 | 之江实验室 | Humanoid piano playing robot |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113008234B (en) * | 2021-02-09 | 2023-04-18 | 北京智能佳科技有限公司 | Group cooperation system based on indoor positioning device |
CN114495959A (en) * | 2021-12-14 | 2022-05-13 | 科大讯飞股份有限公司 | Voice enhancement method, device and system |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004286784A (en) * | 2003-03-19 | 2004-10-14 | Yamaha Corp | Apparatus and program for automatic play of music |
CN101271318A (en) * | 2007-03-21 | 2008-09-24 | 明基电通股份有限公司 | Interactive family entertainment robot and relevant control method |
CN102314917A (en) * | 2010-07-01 | 2012-01-11 | 北京中星微电子有限公司 | Method and device for playing video and audio files |
CN103095814A (en) * | 2012-12-31 | 2013-05-08 | 合一网络技术(北京)有限公司 | Voice driven close distance sharing method, client-terminal, server and system based on information network |
CN103200300A (en) * | 2012-01-10 | 2013-07-10 | 宇龙计算机通信科技(深圳)有限公司 | Terminal and resource sharing method |
CN103714836A (en) * | 2012-09-29 | 2014-04-09 | 联想(北京)有限公司 | Method for playing audio information and electronic equipment |
CN104978973A (en) * | 2014-10-22 | 2015-10-14 | 腾讯科技(深圳)有限公司 | Audio processing method and device |
CN106446930A (en) * | 2016-06-28 | 2017-02-22 | 沈阳工业大学 | Deep convolutional neural network-based robot working scene identification method |
CN106774814A (en) * | 2015-11-20 | 2017-05-31 | 北京奇虎科技有限公司 | Song sharing method and device in Intelligent worn device motion |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103051794B (en) * | 2012-12-18 | 2014-09-10 | 广东欧珀移动通信有限公司 | Method and device for dynamically setting sound effect of mobile terminal |
US9454604B2 (en) * | 2013-03-15 | 2016-09-27 | Futurewei Technologies, Inc. | Motion-based music recommendation for mobile devices |
CN104135705B (en) * | 2014-06-24 | 2018-05-08 | 惠州Tcl移动通信有限公司 | A kind of method and system according to different scenes pattern adjust automatically multimedia volume |
CN105761742A (en) * | 2014-12-19 | 2016-07-13 | 闫辉 | A music playing pen cap |
CN104883642B (en) * | 2015-03-27 | 2018-09-25 | 成都上生活网络科技有限公司 | A kind of effect adjusting method |
US9690817B2 (en) * | 2015-09-01 | 2017-06-27 | International Business Machines Corporation | Song selection using a heart rate change and a facial expression monitored with a camera |
CN107332994A (en) * | 2017-06-29 | 2017-11-07 | 深圳传音控股有限公司 | A kind of tuning effect Self Matching method and system |
-
2017
- 2017-12-12 CN CN201711322453.6A patent/CN108053815A/en active Pending
- 2017-12-21 WO PCT/CN2017/117681 patent/WO2019114015A1/en active Application Filing
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004286784A (en) * | 2003-03-19 | 2004-10-14 | Yamaha Corp | Apparatus and program for automatic play of music |
CN101271318A (en) * | 2007-03-21 | 2008-09-24 | 明基电通股份有限公司 | Interactive family entertainment robot and relevant control method |
CN102314917A (en) * | 2010-07-01 | 2012-01-11 | 北京中星微电子有限公司 | Method and device for playing video and audio files |
CN103200300A (en) * | 2012-01-10 | 2013-07-10 | 宇龙计算机通信科技(深圳)有限公司 | Terminal and resource sharing method |
CN103714836A (en) * | 2012-09-29 | 2014-04-09 | 联想(北京)有限公司 | Method for playing audio information and electronic equipment |
CN103095814A (en) * | 2012-12-31 | 2013-05-08 | 合一网络技术(北京)有限公司 | Voice driven close distance sharing method, client-terminal, server and system based on information network |
CN104978973A (en) * | 2014-10-22 | 2015-10-14 | 腾讯科技(深圳)有限公司 | Audio processing method and device |
CN106774814A (en) * | 2015-11-20 | 2017-05-31 | 北京奇虎科技有限公司 | Song sharing method and device in Intelligent worn device motion |
CN106446930A (en) * | 2016-06-28 | 2017-02-22 | 沈阳工业大学 | Deep convolutional neural network-based robot working scene identification method |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108922315A (en) * | 2018-08-21 | 2018-11-30 | 福州大学厦门工艺美术学院 | A kind of guitar auxiliary practice device |
CN110136677A (en) * | 2019-03-28 | 2019-08-16 | 深圳芒果未来教育科技有限公司 | Musical sound control method and Related product |
CN110136677B (en) * | 2019-03-28 | 2022-03-15 | 深圳市芒果未来科技有限公司 | Musical tone control method and related product |
WO2021068812A1 (en) * | 2019-10-12 | 2021-04-15 | 平安科技(深圳)有限公司 | Music generation method and apparatus, electronic device and computer-readable storage medium |
WO2024008217A1 (en) * | 2023-06-08 | 2024-01-11 | 之江实验室 | Humanoid piano playing robot |
Also Published As
Publication number | Publication date |
---|---|
WO2019114015A1 (en) | 2019-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108053815A (en) | The performance control method and robot of a kind of robot | |
CN111512359A (en) | Modular automatic music production server | |
CN104040618B (en) | For making more harmonious musical background and for effect chain being applied to the system and method for melody | |
KR101326897B1 (en) | Device and Method for Providing a Television Sequence | |
JP2021516787A (en) | An audio synthesis method, and a computer program, a computer device, and a computer system composed of the computer device. | |
CN101657816B (en) | Web portal for distributed audio file editing | |
US20110126103A1 (en) | Method and system for a "karaoke collage" | |
KR20060069209A (en) | Hand portable device and method for playing electronic music | |
CN101103386A (en) | System and method for music score capture and synthesized audio performance with synchronized presentation | |
JP2003076380A (en) | Method of displaying videos of users' own making as karaoke sing-along background videos with karaoke sing- along machines scattered in various places | |
CN110309327A (en) | Audio generation method, device and the generating means for audio | |
EP2396788A2 (en) | Device and method for playback pacing control of a file of signals to be reproduced | |
JP2014085644A (en) | Karaoke system | |
CN106910491A (en) | Karaoke OK system | |
CN106484856A (en) | Audio frequency playing method and device | |
CN110634465A (en) | Music matching method, mobile terminal, data processing method and music matching system | |
CN104822095A (en) | Composite beat special effect system and composite beat special effect processing method | |
CN201118925Y (en) | A microphone four sound control Kara OK song name | |
CN101370216B (en) | Emotional processing and playing method for mobile phone audio files | |
JP2013156543A (en) | Posting reproducer and program | |
JP6058991B2 (en) | Singing video selection system corresponding to singing voice | |
CN104822094A (en) | Polyrhythm special-effect system and polyrhythm special-effect processing method | |
JP2015041091A (en) | Communication karaoke system having feature in switching processing of voice and the like during communication duet | |
CN110415677B (en) | Audio generation method and device and storage medium | |
JP5472245B2 (en) | Karaoke equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180518 |