JP4154600B2 - Automatic accompaniment generation program and automatic accompaniment generation device - Google Patents

Description

The present invention relates to an automatic accompaniment generating program and an automatic accompaniment generating apparatus for causing a computer suitable for performing automatic accompaniment using accompaniment style data in an electronic musical instrument or the like.

  Conventionally, in an electronic musical instrument, a sequencer, and the like, an accompaniment style is used when an automatic accompaniment is added to music data for playback. The accompaniment style includes a plurality of types of sections such as an intro section used for the introduction of the music, a main section used mainly in the music main body, and an ending section used for the ending of the music. As a technique for adding such accompaniment style to music data, there are those disclosed in Japanese Patent No. 3327188 and Japanese Patent No. 2806351.

In Japanese Patent No. 3327188, chord progression chord strings and accompaniment style section strings are automatically generated by analyzing song data. Japanese Patent No. 2806351 can determine a variation type of a section when analyzing song data and automatically detecting a section row, and the automatically determined variation type corresponds to the corresponding section. It corresponds to the climax tendency in the song data of the section to be.
Japanese Patent No. 3327188 Japanese Patent No. 2806351

  When automatic chord progression chord progressions and accompaniment style section sequences are automatically added to song data as in the prior art, conventionally, the entire song data or the entire specific track is recognized as melody data for one song. It was. For example, if there is an intro phrase first and the song data that continues to the actual melody phrase does not have a phrase similar to the intro phrase in the melody, the rise of the section string detected in the melody section The trend may be inappropriate. For example, the intro phrase may have a flashy accompaniment, and instead a quiet accompaniment may occur where it should be swelled by a melody.

  Also, if the intro phrase is on a separate track from the melody, analyzing only the melody as data for one song will not inconvenience the accompaniment trend detected for this melody, No chord string or section string is added to the intro phrase.

  It is an object of the present invention to add an automatic accompaniment suitable for music data without requiring a special operation when performing an automatic accompaniment.

The automatic accompaniment generation program according to claims 1 and 2 is executed by a computer in a personal computer, an electronic musical instrument or the like. By executing the automatic accompaniment generating program according to claim 1, event information is read from the song data, and the performance data in the song data is divided into a plurality of section groups based on the event information, and each of the divided section groups is divided. The performance data was analyzed independently for each section group to generate automatic accompaniment such as chord progression chord progressions and accompaniment style section sequences.

  When dividing performance data into section groups based on event information, a method of detecting, for example, a melody part from event information that is the performance data itself, and making this melody part correspond to one section group, for example, the start position of the melody part There is a method in which event information, which is delimiter information indicating the end position and the end position, is recorded in advance in performance data, and the melody portion is associated with one section group by this delimiter information (event information).

  A song is a series of phrases with different characteristics, such as intro, melody, interlude, and ending, in chronological order. The song data is divided into sections on the time axis, and the performance data (notes) in that section Event) are grouped into section groups, and automatic accompaniment is generated for each section group based on the performance data, it is possible to add an automatic performance suitable for each phrase with phrases having different characteristics.

The automatic accompaniment generator according to claim 1 is directed to a music data comprising a plurality of tracks, for each of the divided segment groups were in generating an automatic accompaniment in each segment groups independently a original accompaniment generation performance The combination of tracks from which data is read is controlled for each section group.

  For example, in the section group corresponding to the intro, performance data is read from tracks other than the melody track. In the section group corresponding to the melody, the performance data is read from the melody track, or the performance data is read from the melody track and the interlude track.

The automatic accompaniment generation program according to claim 2 targets music data composed of a plurality of tracks, and generates each of the divided accompaniments independently for each section group for each section group as in claim 1. It is determined whether there is performance data equivalent to a melody in a section group. If it is determined that there is performance data equivalent to a melody, an automatic accompaniment is generated for that section group based on the performance data of the melody track. When it is determined that there is no performance data equivalent to the melody , the data of each section is analyzed for the section group based on the performance data of tracks other than the melody track, and an automatic accompaniment is generated.

  For example, the melody output part is the performance data of the melody track, but this output part is not “performance data equivalent to the melody”, and automatic accompaniment can be generated based on the performance data of the intro track other than the melody track. Automatic accompaniment suitable for intro can be generated.

According to the automatic accompaniment generator of claim 3, the same effect as that of claim 1 is obtained.

According to the execution of the automatic accompaniment generating program according to claims 1 and 2 of the present invention or the automatic accompaniment generating apparatus according to claim 3 , the music data is automatically analyzed to separate at least a melody phrase section and other phrase sections. Therefore, an automatic accompaniment can be generated independently for each section group, so that an appropriate automatic accompaniment can be added to a phrase in any section in the music data. Since the user does not need to check the phrase structure in the song data by himself / herself, no complicated work is required.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a hardware configuration block diagram of an electronic musical instrument to which the present invention is applied. The CPU 1 controls the entire electronic musical instrument using a working area of the RAM 3 based on a control program stored in the ROM 2. As a basic function of a normal electronic musical instrument, the CPU 1 detects an operation event such as a keyboard of the input operation unit 4 and controls manual performance. Further, an operation event such as a panel operator of the input operation unit 4 is detected, and processing corresponding to the operation of various operators is performed. Further, the CPU 1 controls display on the display 6 via the display unit 5.

  The sound source 7 performs time-division multiplexing processing on a plurality of sound source channels so that a plurality of sounds can be generated substantially simultaneously. A key code, tone data, volume data, etc. are set for each sound source channel from the CPU 1. A tone signal is output to the sound system 8 for each sound source channel. The sound system 8 performs D / A conversion, amplification, etc. of the musical sound signal and generates sound through a speaker or the like.

  The external storage device 9 is a hard disk device (HDD), a floppy (trademark) disk device (FDD), a CD-ROM device, a magneto-optical disk (MO) device, a digital multipurpose disk (DVD) device or the like. 9 is used to supply the automatic accompaniment generation program, song data, automatic accompaniment style data of the present invention from 9 and to store various data. Note that preset accompaniment style data and demo song data are stored in the ROM 2.

  Furthermore, various data is exchanged with an external device 11 such as another MIDI device via a communication interface (I / F) 10. For example, various musical tone control data at the time of reproduction of a song or accompaniment can be output as MIDI data and reproduced by the external device 11. Also, music data and accompaniment style data can be supplied from the external device 11.

  In this embodiment, the control program of the embodiment including the automatic accompaniment generation program, the accompaniment style data other than the song data and the preset are stored in the hard disk device (HDD) of the external storage device 9, and the CPU 1 stores this hard disk. A device control program is developed in the RAM 3 and the operation of the automatic accompaniment is controlled based on the program in the RAM 3. Also, the chord progression chord string and accompaniment style section string are automatically generated by analyzing the song data. Note that the chord string and section string generation method (automatic accompaniment generation) may be a conventional method disclosed in the above-mentioned patent documents and the like, and detailed description thereof will be omitted.

  FIG. 2 is a diagram showing an example of the format of music data in the embodiment. The song data is, for example, SMF (standard MIDI file) or a format conforming thereto, and is composed of a header, a phrase track, and a master track. In the header, a song name, a song length, and the like are recorded.

  The phrase track is composed of a plurality of tracks (track 1, track 2, track 3,...), For example, each piece of performance data of a melody for track 1, an intro phrase for track 2, and an interlude phrase for track 3. Is recorded. That is, in this embodiment, track 1 corresponds to a “melody track”, track 2 corresponds to an “intro track”, and track 3 corresponds to an “interlude track”. The performance data is composed of time-series note events, and one note event is composed of timing data representing note generation timing and note data (pitch data, velocity data, gate time data).

  The master track also records time signature, tempo, tonality, etc., but the chord progression of chord progression generated by automatic accompaniment generation is recorded on this master track. The chord sequence of chord progression is sequentially recorded as chord events, the chord type and root tone, together with timing data indicating the chord start (or switching) timing. In some cases, the chord progression is recorded on the master track. In such a case, the chord progression may not be generated when the accompaniment is generated. Furthermore, as one feature of the embodiment of the present invention, each piece of delimiter information (position information) of the start position of the melody and the end position of the melody is recorded together with the timing data in the master track corresponding to the performance data of the melody. .

  Note that each timing data described above is location data within a song of an event (for example, note data). For example, the event is the number of beats of what measure (measure) and what number of clocks (beat) ( Clock). Data recorded together with timing data (including delimiter information) is all “event information”.

  FIG. 3 is a diagram showing an example of the format of accompaniment style data in the embodiment. Accompaniment style data includes data for a plurality of styles according to various styles. Each accompaniment style data is composed of a header composed of an accompaniment style name, time signature, tempo, etc., and data of a plurality of sections that give variations to accompaniment patterns such as “Intro A”, “Main A”, and “Ending A”. Yes. The data of each section is composed of setting information including section name, number of measures, setting information of each part (track), etc., and pattern data including note data from which actual accompaniment data is based. For example, the pattern data is composed of track pattern data corresponding to a plurality of parts, and each track pattern data includes note data and timbre data for designating the timbre of each track, for example, performance data having a length of 1 to 2 bars. Timing data representing the generation timing is stored in the order of reproduction. This timing data is also location data in the accompaniment pattern, like the music data.

  FIG. 4 is a diagram conceptually showing an example of song data in the embodiment. Performance data (sequence of events) in each of the tracks 1, 2, and 3 is shown by an elliptical bar corresponding to the elapsed time at the time of playing the song. Is shown. In this example, the intro begins with track 2 (intro track) from the beginning of the song, and the melody begins with an aftert by track 1 (melody track) just before the end of the intro. In the middle of the melody, there is an interlude by track 3 (interlude track), and this interval is a rest interval in the melody track.

  As described above, the delimiter information indicating the start and end of the melody is recorded in advance in the master track of the song data in this embodiment. The melody start delimiter information is recorded immediately after the outcome, and the melody end delimiter information is recorded immediately after the melody end. Then, a section (B section) sandwiched between the delimiter information is a section group mainly centering on the melody track, and for example, the front and back sections (A section) of this section group are mainly sections other than the melody track. Group. That is, the melody output part is included in a section group (section A) centered on other than the melody track, and the interlude phrase is included in a section group (section B) centered on the melody track. In this way, event data (outfact) of the melody track may be included in the section group mainly centering on other than the melody track. Moreover, event data (interlude) other than the melody track may be included in the section group mainly having the melody track.

  As a specific method of dividing into section groups, a list called “event list” (variable group capable of recording any number of data) corresponding to each section group (section A, section B) is used. Store event data corresponding to a group in each event list. Thereby, the performance data in the music data is divided into a plurality of section groups. Then, when generating an actual automatic accompaniment, each event list is sequentially selected, and it is determined whether or not there is performance data (event data) corresponding to the melody in the event list. In response, the source track for automatic accompaniment generation is determined. Note that the performance data of the output part in the melody track is not “performance data equivalent to a melody”.

  5 and 6 are main part flowcharts of the automatic accompaniment generation program executed by the CPU 1, FIG. 5 is a flowchart of the performance data dividing process, and FIG. 6 is a flowchart showing an example of the accompaniment generation process. The process of the flowchart of FIG. 5 is executed according to a mode specified by a main routine (not shown). First, in step S11, music data is read into the RAM 3 to prepare music data, and a target track determination rule is set in step S12. This target track determination rule is a process of setting (selecting) a track used for automatic accompaniment generation, a combination thereof, or a priority order. For example, it is processing such as which track is a melody track and which track cannot be targeted together with the melody track. For example, the determination in FIG. 6 is determined by this setting.

  Next, in step S13, an empty event list for the first section group is prepared, for example, by securing a memory area, and the processing from step S14 onward is performed. In step S14, one event data is read from the music data in time order (timing order), and in step S15, it is determined whether there is valid data. If there is no valid data, it is the end of the music data, and the process proceeds to step S19. If there is valid data, it is determined in step S16 whether the data is delimiter information. If it is not delimiter information, the event data is recorded (added) to the event list, and the process returns to step S14. When this event data is recorded in the event list, track information indicating which track is the event data is also recorded. If it is delimiter information in step S16, an empty event list for the next section group is prepared in step S18, and the process returns to step S14.

  With this processing, for example, in the case of FIG. 4, the event data of the intro phrase of track 2 and the event data of the output portion of track 1 are recorded in the first event list, and when the delimiter information at the start of melody is detected, step S18 is performed. The second event list is prepared, and the event data other than the track 1 output and the event data of the interlude phrase for the track 3 are recorded in the second event list. When the melody end delimiter information is detected, the division process is terminated and the process proceeds to step S19. In this example, there are two pieces of delimiter information, and there are two event lists (section groups). Of course, if there are many pieces of delimiter information, the number of event lists (number of section groups) increases accordingly.

  In step S19, for example, accompaniment generation processing as shown in FIG. 6 is performed. In this process, one event list is selected in step S21 for a plurality of generated event lists, and it is determined whether all event lists have been processed in step S28. The processes in steps S21 to S27 are performed for each event list. Do about the list. As a result of the processing in steps S22 and S23, if there is no event data belonging to the melody track in the event list, or if there is event data belonging to the melody track and all of the event data is output, the process proceeds to step S24. On the other hand, if there is event data belonging to the melody track in the event list and the event data includes data that is not an outcome, the process proceeds to step S25.

  The processing in step S24 corresponds to the case of “no performance data equivalent to melody”, and generates automatic accompaniment based on event data of tracks other than the melody track in the event list. That is, even if there is event data of the melody track, if all of them are output, it is determined that the extracted portion is not performance data corresponding to the melody. Thereby, an automatic accompaniment can be generated without being affected by the output. For example, in the case of FIG. 4, this corresponds to the generation of an automatic accompaniment based only on the event data of the intro phrase of track 2 in section A.

  The processing in steps S25, S26, and S27 is processing in the case of “there is performance data equivalent to a melody”, and event data of a target track (can be a source of automatic accompaniment) along with the melody track in the event list. If there is, an automatic accompaniment is generated based on the event data of that track in the event list and the event data of the melody track. If there is event data for a track that cannot be included in the event list together with the melody track, an automatic accompaniment is generated based only on the event data for the melody track in the event list and the event data for the target track together with the melody track. . For example, in the case of FIG. 4, in the B section, automatic accompaniment is generated with both event data of the melody of track 1 and the interlace phrase of track 3, or the event data of the melody is not used and only event data of the melody is used. To generate an automatic accompaniment.

  In this way, whether or not the track can be targeted together with the melody is set by the target track determination rule (step S12 in FIG. 5). In the case of “there is performance data equivalent to a melody”, an automatic accompaniment may be generated unconditionally based only on event data of the melody track. Such an unconditional condition can also be set by the target track determination rule.

In the embodiment, the process of generating automatic accompaniment based on track 2 (intro track) in section A and automatic accompaniment based on track 1 (melody track) and track 3 (interlude track) in section B is as follows: Claim 1 “corresponding to controlling the combination of tracks for each section group. Also, the section A is automatically generated on the basis of the track 2 (intro track) without considering the output as performance data corresponding to the melody. The processing for generating an automatic accompaniment based on track 1 (melody track) in section B corresponds to claim 2 .

  In the embodiment, the case has been described in which the section information indicating the section group is inserted in advance in the song data. However, for example, the section position of the section group is automatically detected, such as searching for a section having lyrics data corresponding to the melody. You may make it do. In addition, event format delimiter information may be included in the same track as the performance data. Furthermore, instead of the event format delimiter information, delimiter information may be included in the header as setting information indicating delimiter positions.

  In addition, when the section is divided by the event format delimiter information, the delimiter information may be added, deleted, or changed by the user.

  Since the present invention can add an automatic performance suitable for a phrase even when the feeling of the music greatly changes, it is also effective for medley format music data. In addition, it can accurately respond to changes in key.

  The melody track to be processed may be fixed by default or may be set by the user. As for tracks containing performance data other than the melody track, all tracks in the song may be processed, or some of the tracks may be selectable as processing targets.

  It is also possible to connect to a communication network via the communication interface 10 and receive delivery of various data such as the automatic accompaniment generation program, song data or accompaniment style data from a server computer or the like.

  The present invention is not limited to the embodiment, and can be various aspects. For example, the song data is not limited to those input from the outside or created by a performance operation. For example, the user inputs a motif melody or chord progression, and develops the motif melody based on the chord progression, so that the song data ( (Melody) may be generated.

  In the embodiment, the present invention is applied to an electronic musical instrument, but may be configured by a personal computer and software. In addition, the sound source device, the sequencer, the effector, and the like are separate devices, and the devices may be connected using communication means such as MIDI or various networks.

It is a hardware block diagram of an electronic musical instrument to which the present invention is applied. It is a figure which shows an example of the format of the music data in embodiment of this invention. It is a figure which shows an example of the format of the accompaniment style data in embodiment of this invention. It is a figure which shows notionally the example of the music data in embodiment of this invention. It is a flowchart of the performance data division | segmentation process in embodiment of this invention. It is a flowchart which shows an example of the accompaniment production | generation process in embodiment of this invention.

Explanation of symbols

  1 ... CPU, 2 ... ROM, 3 ... RAM, 7 ... sound source, 8 ... sound system, 9 ... external storage device

Claims (3)

Reading event information from song data;
Dividing the performance data in the song data into a plurality of section groups based on the read event information;
Analyzing the data of each section independently for each section group with respect to the performance data of each divided section group, and generating automatic accompaniment;
An automatic accompaniment generation program for causing a computer to execute the,
The song data consists of multiple tracks,
The step of generating the automatic accompaniment generates an automatic accompaniment independently for each section group by controlling, for each section group, a combination of tracks for reading performance data as a source of automatic accompaniment for each divided section group. An automatic accompaniment generating program characterized by being a step of performing Reading event information from song data;
Dividing the performance data in the song data into a plurality of section groups based on the read event information;
Analyzing the data of each section independently for each section group with respect to the performance data of each divided section group, and generating automatic accompaniment;
Is an automatic accompaniment generation program for causing a computer to execute
The song data consists of multiple tracks,
The step of generating the automatic accompaniment determines whether or not there is performance data equivalent to a melody in each of the divided section groups, and if it is determined that there is performance data equivalent to a melody, , Generate automatic accompaniment based on the performance data of the melody track, and if it is determined that there is no performance data equivalent to the melody, generate automatic accompaniment for the section group based on the performance data of the track other than the melody track automatic accompaniment generation program that is characterized in that a step for. Reading means for reading event information from song data;
Dividing means for dividing the performance data in the song data into a plurality of section groups based on the event information read by the reading means;
Automatic accompaniment generating means for analyzing the data of each section independently of the performance data of each section group divided by the dividing means and generating automatic accompaniment;
An automatic accompaniment generator comprising:
The song data consists of multiple tracks,
The automatic accompaniment generating means controls, for each section group, a combination of tracks for reading performance data from which automatic accompaniment is generated for each divided section group, and generates an automatic accompaniment independently for each section group. An automatic accompaniment generator characterized by the above.

Images