JPH1124662A - Musical performance support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a musical performance support device capable of developing a smooth understanding between players and an idea in performance. SOLUTION: A first player plays the first four measures by use of a MIDI (musical instrument digital interface) musical instrument 1, and a second player plays four measures by use of a MIDI musical instrument 2. The sampler 13 of a client 10 samples and outputs the play data from the MIDI musical instrument 1 to a server 5, and also imparts the play data to an MIDI sound source module 4 to produce an actual play sound. The server 5 extracts the profile attribute and color attribute from the play data, and a statistical processing part 54 statistically processes and calculates them as spatial structure information, and displays them on the display part 14, 24 of each client 10, 20.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a music performance support apparatus, and more particularly, to extracting and instructing information related to a performance intention from a performance by using a statistical technique, thereby performing cooperative work among performers particularly in jazz improvisation. And a music performance support device capable of supporting music in real time.

[0002]

2. Description of the Related Art Musical performance by a plurality of players is one of advanced cooperative tasks. Performers communicate with each other to create a consistent performance expression as a whole. Especially in jazz improvisation, the melody to be played is often created from this cooperative work. Non-verbal means such as sounds, gestures, and facial expressions played by each player are usually used as communication means in such music performance venues. However, the expression by these means is ambiguous, and it is not possible to convey specific intentions like a language, and this cooperative work is particularly difficult for beginners.

In recent years, music information processing has been actively researched. Although there are a wide variety of subjects, research that focuses on the aspect of such cooperative work by a plurality of players includes an automatic accompaniment system in which a computer accompanies a human performance. In early automatic accompaniment systems, humans were the main performers, and the main purpose of the system was to follow the tempo of the performance. However, in recent years, an accompaniment system in which the system takes the initiative depending on the performance situation, and an accompaniment system that follows not only the tempo but also the change in the performance expression of the solo player, etc., have been developed. In addition, some systems perform not only accompaniment but also solo performances as a performer equal to humans.

[0004]

The above-mentioned research aims at realizing the cooperative work between man and machine. This is one of the natural trends in the heyday of computer music, but it is unlikely that the creation of music by collaborative work between people will disappear. However, little research has been done on cooperative work between people in music.

[0005] Therefore, a main object of the present invention is to provide a music performance support device which can support cooperative work between people in music performance in real time and further inspire creativity in music performance. is there.

[0006]

The invention according to claim 1 is
Performance data collecting means for collecting performance data of musical instruments provided for each player, and various attribute information extracted from the performance data collected by each performance data collecting means, and statistically processed. It is provided with processing means for calculating as spatial structure information, and display means provided for each player and displaying the spatial structure information calculated by the processing means.

According to a second aspect of the present invention, the processing means of the first aspect extracts a weighted outline attribute and a weighted color attribute from the performance data, and outputs the phrase object of the performance up to the present time and the outline attribute information of each phrase object. And color attribute information, and the relevance between the phrase object and each attribute is calculated as spatial structure information by statistical processing based on all data.

[0008]

FIG. 1 is a block diagram showing an embodiment of the present invention. In FIG. 1, MIDI (Musica
l Instrument digital interface) Clients 10 and 20 are provided corresponding to the musical instruments 1 and 2, respectively. M
The IDI instrument 1 is a master, and the MIDI instrument 2 is a slave. Although only one slave is shown in FIG. 1, two or more slaves may be provided. In that case, a client is provided for each slave.

[0009] Performance data and a clock signal are output from the MIDI musical instrument 1 and input to the client 10, and the clock signal is supplied to the MIDI synthesizing unit 3.
The performance data from the MIDI musical instrument 2 is provided to the MIDI synthesizing unit 3, and a clock signal is added to the performance data from the MIDI musical instrument 2 and the resultant data is provided to the client 20. The client 10 includes a switch 11 and an improvisational performance support device 12.
, A sampler 13 and a display unit 14. The switch 11 switches whether or not to input performance data to the improvisation performance support device 12. The improvisation performance support device 12 supports improvisation performance for beginners. The performance data input from the MIDI instrument 1 or the performance data output from the improvisation performance support device 12 is input to the sampler 13, and the performance data for a certain period is sampled. The sampled performance data is output to the server 5 via the network 6.

[0010] Similarly to the client 10, the client 20 includes a switch 21, an improvisation performance support device 22, a sampler 23, and a display unit 24. Client 1
The performance data 0 and 20 are input to the MIDI tone generator module 4. The MIDI tone generator module 4 automatically plays a prelude.

The server 5 extracts various attributes from the sampling data output to the network 6 and statistically processes them to calculate spatial structure information. To this end, the server 5 includes a music primitive extraction unit 50 and a statistical processing unit 54. The music primitive extraction unit 50 includes a contour attribute extraction unit 51
And a color attribute extracting unit 52 and a memory 53. The contour attribute extraction unit 51 extracts attributes obtained from the time-series movement of the pitch of the sound in the played phrase, and
Reference numeral 2 extracts an attribute based on the emotional effect on the person at each time point depending on the chord of each part of the music. The memory 53 stores the extracted outline attributes and color attributes. The statistical processing unit 54 calculates the relevancy between the phrase object and each attribute as a spatial structure by statistical processing based on the extracted outline attributes and color attributes by dual scaling.

Here, the dual scale method means that when a set of objects composed of complex quantities or attributes is given,
This method expresses the attribute sharing between objects and the co-occurrence between attributes as a relative positional relationship in space by giving a score quantity to each of the object set and the attribute set. In this embodiment, the dual scale method is applied with each phrase as an object, each music primitive as an object attribute, and the weight of the music primitive as an attribute value. This makes it possible to express the relevance of each phrase to each music primitive in one space. The spatial structure information processed by the statistical processing unit 54 is transmitted to each client 1 via the network 6.
0, 20 and displayed on the display units 14, 24.

FIGS. 2 to 4 are flowcharts for explaining a specific operation of the embodiment of the present invention. In particular, FIG. 2 is a flowchart showing a sampling process of a client, and FIG. FIG. 4 is a flowchart showing the operation, and FIG. 4 is a flowchart showing the spatial structure display processing of the client. FIG. 5 is a diagram showing a display example of the spatial structure.

Next, a specific operation of the embodiment of the present invention will be described with reference to FIGS. First, prior to the start of performance, each of the clients 10, 20 and the server 5 sets a music piece to be played. It also sets the total number of performers and the order in which the performers of each client will perform. In this embodiment, it is assumed that the player of the client 10 plays first and the player of the client 20 plays second. Therefore, the performance is Client 1.
0 → client 20 → client 10 → client 20 →...

The performance is started when a clock signal is supplied from the master MIDI musical instrument 1 to each of the clients 10 and 20, and the clients 10 and 20 proceed with the processing in synchronization with the clock signal. Then, the MIDI tone generator module 4 performs an automatic performance of the prelude according to the signal for starting the performance. When the prelude ends, the first player starts playing using the MIDI instrument 1. The first player plays for a predetermined period, for example, only the first four bars, and then the second player plays the next four bars.

When a beginner performs improvisation, the switches 11 and 21 of the clients 10 and 20 can be switched to support the improvisation by the improvisation support devices 12 and 22. In this case, the improvisation performance support device 1
The performance data output from 2, 22 is sampled.

Client 1 corresponding to MIDI musical instrument 1
0 counts the clock signal from the MIDI instrument 1 as shown in FIG. When the client 10 determines that the playing order is its own,
The sampler 13 samples performance data. The sampler 13 samples the performance data for the first four measures and transmits the sampled performance data to the server 5 via the network 6. The performance data input to the sampler 13 is also input to the MIDI sound module 4, and the MIDI sound source module 4 outputs actual performance sounds for the first four bars based on the performance data.

Thereafter, the second client determines that it is in its own performance order, samples the next four bars of performance data, and transmits it to the server 5. Client 2
The performance data input to the 0 sampler 23 is also input to the MIDI tone generator module 4, and the MIDI tone generator module 4 outputs the next four measures of actual performance sounds.

When the server 5 determines that the sampled performance data has been input from the clients 10 and 20 as shown in FIG. 3, the contour attribute extraction unit 51 extracts a weighted contour attribute from the performance data, and The extraction unit 52 extracts a weighted color attribute from the performance data. The statistical processing unit 54 extracts the outline attribute and color attribute information of the current phrase object and each phrase object. Then, the statistical processing unit 54 performs a process of expressing the association between the phrase object and each attribute as a two-dimensional space structure by a statistical processing method called a dual scale method based on all data. The processing result is transmitted to each of the clients 10 and 20 via the network 6. The statistical processing unit 54 stores in the memory 53 the phrase objects of the performance up to the present time, and the outline attribute and color attribute information of each phrase object.

When each of the clients 10 and 20 receives the spatial structure data from the server 5, the corresponding display units 14 and 24
FIG. 5 shows the spatial structure data as shown in FIG.

FIG. 5 shows a display example of a spatial structure at the end of a two chorus (18 phrases) performance in a certain performance, where rectangular icons correspond to individual phrases. The icon of each phrase has a different color for each player, and indicates the name of the player and the number of the ad-lib performance performed by that player. In FIG.
The lighter icon is the performance of the client 10 or 20 user, and the darker icon is the performance of the improvisation performance support device 12 or 22 user. The oval icons correspond to individual music primitives. The music primitives that characterize the first principal component in the horizontal axis direction include an out sound (OUTOFCHORD), a chromatic chord sound (HORDTONE), and an avoid sound (AVOIDNOT).
ES). Characteristic features of the second principal component in the vertical axis direction are a representative repetition length (REPEAT) and a tension sound (TENSION).

[0022]

As described above, according to the present invention, performance data of each player's musical instrument is collected, various attributes are extracted from the collected performance data, and these are statistically processed and visualized as spatial structure information. And display it to each player, so that the sound information of the performance is composed in front of you as spatial information, so you can enjoy the fun, and you can understand the characteristics of your performance well, and moreover, The documents related to the phrases by the players can be seen at a glance, so that it is possible to realize smooth and concrete communication of intentions between the players. Also,
Unexpected phrases and relevance of characteristic attributes may be found from the spatial structure, and it is possible for the player to obtain new discoveries in the creation of performances.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a flowchart illustrating a sampling process of a client.

FIG. 3 is a flowchart illustrating a processing operation of a server.

FIG. 4 is a flowchart showing a spatial structure display process of a client.

FIG. 5 is a diagram showing a display example of a spatial structure.

[Explanation of symbols]

 1, MIDI instrument 3 MIDI synthesizing unit 4 MIDI sound source module 5 Server 10, 20 client 11, 21 switch 12, 22 Improvisation performance support device 13, 23 Sampler 14, 24 Display unit 50 Music primitive extraction unit 51 Outline attribute extraction unit 52 Color attribute extraction unit 53 Memory 54 Statistical processing unit

Continued on the front page (72) Inventor Yasuyuki Kado 5th Sanraya, Inaya, Shiroka-cho, Soraku-cho, Kyoto Pref. ATI Inc. Intelligent Television Image Communication Research Laboratories (72) Inventor Kenji Mase Sera-cho, Soraku-gun, Kyoto 5 inn, small inn, small inn, small inn

Claims (2)

[Claims] 1. A music performance support apparatus for smooth communication between players when a plurality of players play music with musical instruments, and for inspiring ideas in performance, Performance data collecting means for collecting performance data of musical instruments provided for the performer; extracting various attribute information from the performance data collected by each of the performance data collecting means; A music performance support device, comprising: processing means for calculating as structural information; and display means provided for each of the performers and displaying the spatial structural information calculated by the processing means. 2. The processing means according to claim 1, wherein the processing means includes a weighted contour attribute based on time-series motion information of a pitch of a sound in the phrase from the performance data, and a pitch at each time point depending on a chord of each part of the music. Weighted color attributes based on the affective effect on the phrase object, extract the phrase objects of the performance up to the present and the outline attribute information and color attribute information of each phrase object, and perform statistical processing based on all data to identify the phrase object The music performance support apparatus according to claim 1, wherein the relevance of each attribute is calculated as spatial structure information.