JP3532485B2 - Editing method, apparatus, and program recording medium for reducing the number of channels of electronic music data - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a predetermined number of electronic musical score type music data in which musical performance information of a large number of musical instrument parts constituting a musical piece is expressed by musical score electronic data of a large number of channels, for example, MIDI standard musical performance data. It is related to the technology to edit the music data that has been reduced and edited in the channel.

[0002]

2. Description of the Related Art MIDI (Musical Instrument Digital)
Interface) music data is electronic musical score type music data in which musical performance information of a large number of musical instrument parts constituting a musical composition is represented by electronic musical score type performance data of a large number of channels. A tone generator that reproduces MIDI data can simultaneously generate the tone colors of a number of musical instrument parts corresponding to the number of channels that the tone generator has, and a tone generator with a larger number of channels has a tone color that corresponds to more musical instrument parts. Can be pronounced simultaneously. Usually, the tone generator device stores tone color data for hundreds of musical instrument parts, and the MIDI data includes information indicating which tone color data is assigned to which channel.

Recently, MIDI data has become established as a de facto standard as a standard data format for exchanging music data through personal computer communication or the Internet. In addition, a sound source device such as a sound source board is now standardly installed in a personal computer, and the number of lovers who enjoy MIDI data downloaded through a CD-ROM attached to a personal computer magazine or personal computer communication or the Internet is rapidly increasing.

[0004]

The present applicants, paying attention to the above situation, plan a service for distributing karaoke music data in the MIDI data format through personal computer communication or the Internet as part of the karaoke business.
In order to operate this service efficiently, we are considering using a huge number of karaoke music data created over many years in the online karaoke system business.

However, there is a problem in reproducing karaoke data for a communication karaoke system on a personal computer.
The online karaoke performance terminal is equipped with a high performance sound source device of 32 to 64 channels. Therefore, the karaoke music data is a set of performance data of 32 to 64 channels. In order to play this as it is on a personal computer equipped with a sound source board of about 16 channels, the number of channels must be reduced by some method. For example, when reproducing karaoke music data of 64 channels with a tone generator board of 16 channels mounted on a personal computer, the performance data of the Nth channel after the 17th channel is the remainder value of "N ÷ 16". Is assigned to the channel corresponding to. Therefore, a plurality of performance data will be assigned to one channel in duplicate, resulting in performance data that is not played with the original tone color, and sometimes the tone color is completely different from the tone color that the original karaoke music data assumed. It may happen that a tone is assigned.

Such a problem also occurs when lovers exchange MIDI data by personal computer communication or the Internet. For example, the number of channels of the sound source device used by the creator of the MIDI data depends on the user. It is often the case that the number of the sound source devices used is larger than that of the sound source devices used, but the same problem as described above may occur in this case as well.

The present invention has been made in view of the above circumstances, and an electronic musical score in which the musical performance information of a large number of musical instrument parts is expressed by the musical score data of a large number of channels while maintaining the performance performance as high as possible. It is an object of the present invention to provide a method for editing formula music data into music data reduced to a predetermined number of channels.

[0008]

The invention according to the first aspect of the present invention which achieves this object is a method of reducing the number of channels of electronic musical score type music data, which composes music. For electronic musical score type music data in which performance information of many musical instrument parts is represented by electronic musical score type performance data of many channels, the total pronunciation time of the musical performance data of each channel is calculated, and a predetermined number In addition to selecting the performance data of a channel (this is the performance data of the selected channel), based on the similarity of the tone color data assigned to each channel, the unselected channels (this is the performance data of the unselected channel). The selected channel is similar, and the performance data of the selected channel contains the performance data of another non-selected channel that is similar to the tone. By be mixed, and to obtain music data having a reduced edited to the predetermined number of channels.

The invention according to the second aspect of the present invention is not selected when the performance data of another channel similar in tone color is mixed into the performance data of the selected channel in claim 1. The value of the parameter data that specifies the amount of change in the pitch bender included in the performance data of the channel is changed to a value based on the pitch bend range of the performance data of the selected channel that is the mixing destination.

According to a third aspect of the present invention, in the first or second aspect, the selected channel includes a channel to which tone color data of at least one rhythm sound is assigned. I decided to.

The invention according to a fourth aspect of the present invention is a recording medium recording a program for executing the reduction editing method according to any one of the first to third aspects by a computer. To do.

Further, a fifth aspect of the present invention is a programmed computer for carrying out the method according to the first aspect of the present invention, wherein the performance information of a large number of musical instrument parts constituting a musical composition. Means for reading the electronic musical score source music data expressing the electronic musical score type performance data of a large number of channels, means for obtaining the total pronunciation time of the musical performance data of each channel of the source music data, and A means for registering performance data of a predetermined number of channels in descending order as non-selected channels, and a non-selected channel A
Means for determining one selected channel B having the most similar tone color to the non-selected channel A based on the similarity between the tone color data assigned to the selected channel and the tone color data assigned to each selected channel; A means for mixing the performance data of the non-selected channel A in the performance data, and a means for outputting object music data consisting of a predetermined number of performance data of the selected channel in which the performance data of the non-selected channel is appropriately mixed. An editing apparatus for reducing the number of channels of electronic musical score type music data, which is characterized by including the following.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION MI for a communication karaoke system created with a 64-channel structure which is source music data.
An embodiment of the present invention will be described below by taking as an example a method of converting the DI data A into MIDI data B of a sound source device of 16-channel structure which is object music data.

In this embodiment, when the source music data is converted into object music data, 16 channels having the longest total pronunciation time are selected from the source music data as selected channels, and these selected channels are assigned in the source music data. The assigned musical instrument part can be assigned as it is in the object music data. The selection of the selected channel according to the total pronunciation time is based on the experience that the sound of the instrument part with a longer total pronunciation time is more likely to remain in the ear. If the parts are played, it is possible to prevent deterioration of performance quality to a minimum. Note that the total pronunciation time is calculated based on the time from the start to the end of the pronunciation of a certain note in a normal musical score, for example, Δtime data that determines the note-on event to the note-off event in the source music data. To do.

Regarding the performance data of the non-selected channels other than the selected channel, the instrument parts (timbre data) assigned to the respective non-selected channels are compared with the instrument parts assigned to the selected channel to unselect them. Mix the instrument part that most closely resembles the instrument part of the channel with the performance data of the selected channel, that is, the performance data of the non-selected channel is adopted as the performance data of the instrument part with a similar tone color. There is. As a result, the performance data of the non-selected channel is reflected in the object music data, and is mixed with the performance data of the musical instrument part similar to the tone color, so that the deterioration of the performance quality can be minimized.

=== Conversion Process === Next, the conversion process will be specifically described. MIDI
The conversion from the data A to the MIDI data B is performed by a conversion program running on a computer such as a personal computer or a workstation equipped with a CPU, a memory, a hard disk device, a CD-ROM device and the like. The structure of MIDI data will be briefly described. MI as shown in FIG.
DI data consists of two types of blocks, a header block 3 and a track block 5. In the header block, information about the structure of MIDI data such as the number of track blocks is described. On the other hand, in the track block 5, message numbers corresponding to control of musical instruments such as note-on events and note-off events and their control targets are recorded. MIDI event 7 which is a channel message in which a set channel ID is a set, exclusive event 9 such as Δ time data described above, correspondence information between each channel and audio data, and musical score information such as tempo / beat / key of music. , A lyric information, a meta event that is information such as a singer name and a composer name, and a meta event 11 that is data related to various information are described.

FIG. 2 is a flow chart for explaining the processing of the conversion program. First, the (100) conversion program started in response to an operator's double-click or the like converts the MIDI data A of the conversion source corresponding to the file name given as a parameter at the time of the operator's key input operation or startup from the hard disk. Read and load into memory (1
10). And the conversion program is the M read in the memory
Based on the channel data described in the track block 5 of the IDI data and the Δtime data attached to the channel data, the total pronunciation time of the musical tone for each channel is calculated (12
0) to determine 16 channels in descending order of total pronunciation time, and assign the channel ID to the MID on the memory.
The data is written in an area different from the I data (130). Here, for example, for so-called drum-type instrument parts such as drum sets, cymbals, hi-hats, and snares included in karaoke music data, if the performance data of each instrument part is played by another instrument part, a completely different music In some cases, the performance may be unbearable or unbearable. Therefore, for the MIDI data A including these musical instrument parts, the channels to which these musical instrument parts are assigned are not targeted for reduction and are preferentially adopted as the selected channels.

Next, the conversion program refers to the channel ID of the selected channel written in the memory, and for each non-selected channel, the tone color data most similar to the tone color data assigned to each of these non-selected channels. Determines the selected channel to which is assigned. Then, the determined correspondence is stored in the memory (140). At this time, which tone color data is assigned to each channel is investigated based on the correspondence information between each channel and audio data described in the meta event 11 described above. Further, the degree of similarity is determined according to the similarity table shown in FIG. 3 which is distributed together with the conversion program and stored in the hard disk, and two tone color data having higher values in this table are determined to be more similar. In this similarity table, the type of each tone color data is indicated by a tone color number (for example, conforming to the GM standard). For example, the similarity between the piano and the jazz piano is similar, and therefore the similarity is high, and the similarity between the piano and the guitar is lower than this.

After checking the selected channel to which the most similar tone color data is assigned to each of the non-selected channels in this manner, the correspondence relation stored in the memory is then referred to and loaded into the memory. The channel ID of the channel message of the non-selected channel in each MIDI data A is set to the channel ID of the selected channel associated with each non-selected channel.
(150). As a result, the performance data of the non-selected channel is mixed with the performance data of the selected channel associated with the performance data. At this time, for non-selected channels with extremely short total pronunciation time,
If this is purposely mixed in the performance data of the selected channel, the performance quality may be deteriorated.
Therefore, an algorithm for discarding the performance data of the selected channel without mixing it with the performance data may be incorporated in the conversion program.

Then, when the conversion process is completed for all the non-selected channels, the conversion program determines the MID based on the changed MIDI data in the memory.
A file in which I data is described is created on the hard disk (160). As described above, a file in which the MIDI data B of 16-channel structure which is the object music data is described is created.

=== Others === By the way, MIDI data contains a channel message called a pitch bend message for producing playing effects such as guitar choking and saxophone hiccup feeling. Parameter data that specifies the pitch bend message may be mixed, and the MIDI data in which the pitch bend message is mixed in this way always has the pitch bend sensitivity (or pitch bend sensitivity) that indicates the minimum and maximum pitch bend widths for each channel. It contains information called a range. When mixing the performance data of the non-selected channel into the performance data of the selected channel by the conversion program described above, if the pitch bend ranges of the two are different, this may cause discontinuity or discomfort in the performance. The quality will be significantly degraded. Therefore, when the performance data of the non-selected channel is mixed into the performance data of the selected channel, if the pitch bend message is included in the performance data of the non-selected channel, the value of the parameter data is mixed into the performance of the selected channel. An algorithm for correcting the pitch bend width of the data to a value with the standard scale may be incorporated in the conversion program.

Although the similarity table may be distributed together with the conversion program as described above, only the similarity table may be distributed independently. In this way, for example, a better performance can be achieved. It is possible to upgrade to a similarity table that creates MIDI data that can be played back. It is also interesting that the conversion program is provided with a function that allows the operator of the conversion program to customize the similarity table so that the user of the MIDI data can edit the similarity table for organizing the MIDI data that they are satisfied with. .

Further, an optimum similarity table may be prepared for each music genre so that the optimum conversion can be performed for each music genre.

[0024]

As described above, according to the present invention,
Performance data of the main instrument parts of the music is preferentially adopted as the edited music data, and performance data of unselected channels are also mixed into the performance data of the selected channel that is similar to the timbre. Editing electronic musical score music data in which musical performance information of many musical instrument parts is expressed by electronic musical score musical performance data of many channels while reducing the edited music data to a predetermined number of channels while maintaining high performance quality. You can

[Brief description of drawings]

FIG. 1 is a diagram showing a schematic data structure of MIDI data according to an embodiment of the present invention.

FIG. 2 is a diagram showing a flowchart showing processing of a conversion program according to an embodiment of the present invention.

FIG. 3 is a diagram showing a similarity table according to an embodiment of the present invention.

[Description of Codes] A process of calculating the total pronunciation time for each of 120 channels. 130 Processing for selecting performance data of a predetermined number of channels in descending order of total pronunciation time. 140 Processing of determining a selected channel to which a non-selected channel is similar, based on the degree of similarity of tone color data. 150 Processing for mixing performance data of a non-selected channel similar to a tone color into performance data of a selected channel.

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G10H 1/00-7/12 G10K 15/04 302

Claims (5)

(57) [Claims] 1. For electronic musical score type music data in which musical performance information of a large number of musical instrument parts constituting a musical piece is expressed by electronic musical score type musical data of a large number of channels, a total pronunciation time of the musical performance data of each channel is obtained. The performance data of a predetermined number of channels are selected in descending order of total sound generation time (this is the performance data of the selected channel), and they are not selected based on the similarity of the tone color data assigned to each channel. The selected channel that has a similar channel (this is the performance data of the non-selected channel) and mixes the performance data of another non-selected channel similar to the timbre into the performance data of the selected channel, and Reducing the number of channels Reducing the number of channels for electronic musical score music data, characterized by obtaining edited music data How to edit. 2. The amount of change in pitch bender included in the performance data of a channel not selected when the performance data of another channel similar to the tone color is mixed into the performance data of the selected channel according to claim 1. A method for reducing the number of channels, characterized in that the value of the parameter data to be changed is changed to a value based on the pitch bend range of the performance data of the selected channel to be mixed. 3. The reduction of the number of channels of electronic musical score music data according to claim 1, wherein the selected channel includes a channel to which tone color data of at least one rhythm sound is assigned. Method. 4. A recording medium recording a program for executing the reduction editing method according to claim 1 by a computer. 5. A computer programmed to carry out the method according to claim 1, wherein performance information of a large number of musical instrument parts constituting a musical composition is represented by electronic musical score performance data of a large number of channels. A means for reading the electronic musical score type source music data, a means for obtaining the total pronunciation time of the performance data of each channel of the source music data, and a performance data of a predetermined number of channels is registered as a selected channel in descending order of the obtained total pronunciation time. In addition, the non-selected channel A is the most selected one based on the similarity between the tone color data assigned to a certain non-selected channel A and the tone color data assigned to each selected channel. Means for determining one selected channel B similar in tone color, and performance data of the selected channel B And means for mixing the performance data of the non-selected channel A, and means for outputting object music data consisting of a predetermined number of performance data of the selected channel into which the performance data of the non-selected channel is appropriately mixed. A device for reducing the number of channels of electronic musical score type music data, which is characterized in that