JP4211650B2 - Accompaniment pattern generation apparatus and accompaniment pattern generation program - Google Patents

Description

  The present invention relates to an accompaniment pattern generation device and an accompaniment pattern generation program.

Conventionally, there are apparatuses that automatically generate an accompaniment pattern or edit an accompaniment part.
For example, if you select a melody from the music database, specify the tempo, and enter a kansei expression that matches the tune's style, you can create an accompaniment pattern consisting of multiple tones that match the melody's sensibility. There is an automatic generation device. (See Patent Document 1)
In addition, there is a device for changing the combination of each performance section of the original accompaniment part, arranging the original accompaniment part, and editing it as a new accompaniment part. (See Patent Document 2 and Patent Document 3)

Japanese Patent Laid-Open No. 10-105172 (first to sixth pages, FIGS. 1 to 8) JP 2002-229561 A (pages 1 to 20 and FIGS. 1 to 12) JP 2003-216156 A (1st page to 9th page, FIGS. 1 to 6)

The conventional accompaniment pattern automatic generation device uses a technique of selectively specifying keywords of expression that match the melody part's musical style according to the user's own sensibility. There was a tendency for accompaniment parts to be generated due to the musical sensibility of the traditional concept, and it was apt to be a song lacking individuality and novelty.
In addition, since the conventional performance data editing apparatus simply replaces and edits any data of an existing accompaniment part or changes the chord data of an existing accompaniment part to another chord data structure, There was a problem that it was not possible to deviate from the fixed concept.

In the first aspect of the invention, the waveform data capturing means for capturing the waveform data, the pattern data storage means for storing the timing pattern data of the sounding and muting of the specified tone color waveform, and the pattern data storage means are stored. Data stored in the standard tone color storage means, and standard tone color data storage means for storing data representing the characteristics of the tone color specified by the timing pattern data, and the waveform data characteristics acquired by the waveform data acquisition means When the judgment means for judging whether or not the waveform data captured by the judgment means is similar, the timing pattern stored in the pattern data storage means By assigning waveform data as the specified tone waveform, accompaniment patterns Characterized in that it has a pattern generating means for generating a down, a.
With such a configuration, if the characteristics of the sample waveform data are similar to the characteristics of the standard timbre pattern, it can be added to the memory as new timbre waveform data. By using these tone color waveform data, it is possible to generate a musical tone of accompaniment rhythm with an unprecedented novel tone color.

The pattern data storage means according to claim 2 further stores a plurality of pattern data corresponding to different timbres, and the standard timbre data storage means stores data representing the characteristics of the different timbres. Further, the apparatus has selection means for selecting one pattern data from a plurality of pattern data stored in the pattern data storage means and for selecting data representing timbre characteristics corresponding to the selected pattern data. May be.

  The waveform data acquisition means according to claim 3 may acquire desired waveform data from a waveform data storage means in which a plurality of types of waveform data are stored.

  The waveform data capturing means according to claim 4 may be connected to an external database for storing desired waveform data to capture the desired waveform data.

  As described in claim 5, the present invention can be implemented by programming the operation procedure of each means described above and causing a computer to execute it. At that time, the computer program can be supplied to the computer through a disk-type recording medium such as a hard disk, various memories such as a semiconductor memory or a card-type memory, or various program recording media such as a communication network.

The accompaniment pattern generation apparatus of the present invention adds the waveform data to the memory as new timbre waveform data if the characteristics of the waveform data specified as the sample are similar to the characteristics of the standard timbre data. Waveform data other than the existing timbre can be added to the database and used effectively as a variety of musical timbres.
Further, by adopting such a configuration, even a user who does not have musical knowledge selects timbre waveform data that seems to be optimal from the timbre waveform data database and uses it. Since it can be assigned to a timbre pattern desired by the user, there is also an effect that a rhythm accompaniment with a novel and unique timbre can be enjoyed.

If the characteristics of the sample waveform data are similar to the characteristics of the standard timbre pattern, this is taken into memory as new timbre waveform data and assigned to the timbre pattern as a template based on these timbre waveform data. It is to generate rhythm music.
Example 1

FIG. 1 shows a block diagram of the overall configuration of the present invention in an embodiment of the present invention.
The accompaniment pattern generation apparatus of the present embodiment is taken as an example when applied to an electronic keyboard instrument.
The CPU 101 is connected to each circuit and device, which will be described later, constituting the accompaniment pattern generation device via the bus line 111, and controls various functions by exchanging various data.

  The program ROM 102 stores a program for the CPU 101 to execute processing of various functions according to a predetermined procedure.

The work RAM 103 is a work area that includes a plurality of registers and temporarily stores various data in the course of data processing.
The data area of the data memory 104 includes an area of a data group serving as a template for an accompaniment pattern, an area of a rhythm data group, an area of standard tone color data, and a data group of search tone waveform corresponding to the template data.

The waveform memory 105 is generated by electronic means such as waveform data of natural phenomena such as wind and wave sounds, waveform data of sounds of humans and animals, waveform data generated by natural instruments, or electronic musical instruments. A wide variety of waveform data such as waveform data is stored.
The musical sound generation device 106 is provided with an electronic keyboard instrument, for example.
This electronic keyboard instrument is composed of a keyboard, a tone generator circuit, a sound generation circuit, and the like, and generates and generates musical sounds based on event data or a signal corresponding to a keyboard operation.
The operator group 107 includes a plurality of switches, and includes a power switch, a mode switch, a switch operated during rhythm creation processing, a switch operated during rhythm reproduction processing, and the like.

The display device 108 is composed of, for example, a liquid crystal display device, and displays various screens during rhythm creation processing and rhythm reproduction processing.
The communication circuit 109 is an interface circuit for reading and supplying sample waveform data from the external database 110.
The waveform data stored in the external database 110 is waveform data having characteristics different from the characteristics of the waveform data stored in the waveform memory 105.

2 and 3 show the data structure in the data memory 104 shown in FIG.
The data memory 104 includes areas of a template data group, a rhythm data group, a standard tone color data group, and a search tone color data group.
First, the area of the template data group shown in FIG. 2 is made up of template data (0) to template data (N) for accompaniment in which a plurality of types of rhythm instruments such as percussion instruments are organized. (N) consists of areas of timbre (1) pattern to timbre (n) pattern corresponding to the timbre of each instrument.
In the timbre (1) pattern area, event (on) on-identification data having a sound generation start as a timing pattern and event (off) off-identification data as a timing pattern for instructing to mute the sounded sound are displayed. Is stored.
Next, the rhythm data group includes a data area of rhythm data (0) to rhythm data (M).
For example, the area of rhythm data (0) is composed of a timbre (1) pattern to a timbre (m) pattern, and the timbre (1) pattern area is composed of an on-event area and an off-event area.
In this on-event area, identification data for turning on pronunciation, a start address indicating the start position of the waveform data stored in the search tone data group, and an end address data indicating the end position of the waveform data are stored. Is done.
In the off event area, identification data for turning off pronunciation is stored.

Next, as shown in FIG. 3, the standard timbre data area is composed of standard timbre (0) to standard timbre (P) areas. For example, the standard timbre (0) area includes timbre frequency characteristic data. Envelope data and pitch data are respectively stored.
Next, the area of the search timbre data group consists of areas of the search timbre (0) to the search timbre (S). For example, the search timbre (0) area corresponds to the timbre pattern of the template data group. Tone color waveform data AD1 to ADs + 1 are stored.

Next, the operation of the basic processing of the present invention will be described with reference to FIG.
First, when a power switch provided in the operator group 107 is turned on, the CPU 101 initializes each circuit of the entire accompaniment pattern generation apparatus (S401).
Next, “0” is set to the mode flag MODE (S402).
Then, it is determined whether or not the mode flag MODE is “0” indicating the rhythm creation process (S403).
Here, if MODE indicates “0”, rhythm creation processing is executed (S404), and if MODE does not indicate “0”, rhythm reproduction processing is executed (S405).
When the rhythm creation process and the rhythm reproduction process are finished, the process returns to S403 again.

<About the rhythm creation process>
The detailed operation of the rhythm creation process (S404) in FIG. 4 will be described with reference to FIGS.
First, an initial screen for creating a rhythm is displayed on the display device 108 (S501).
A subroutine for selecting a template of an accompaniment tone color pattern is executed based on the guidance display for creating the rhythm (S502), and then a subroutine for searching for a waveform corresponding to the tone pattern of the selected template. The process is executed (S503).
Then, the subroutine processing for assigning the retrieved waveform to the template tone color pattern is executed (S504).

Next, when the subroutine processing is completed, it is determined whether or not the mode switch is turned on (S505).
If it is detected that the mode switch is turned on, the mode flag MODE is set to “1” indicating the rhythm reproduction process, and the process is terminated (S506).
On the other hand, if it is not detected that the mode switch is turned on, the process returns to S501 again.

<About template selection processing>
Next, the detailed operation of the template selection process (S502) in FIG. 5 will be described with reference to FIG.
First, it is determined whether or not a switch for selecting any template of the accompaniment timbre pattern provided in the operator group 107 is turned on (S601).
When it is detected that the template selection switch is turned on, a screen for selecting a template of the accompaniment tone color pattern is displayed on the display device 108 (S602).
On the other hand, if it is not detected that the template selection switch is turned on, this process ends.

Next, based on the displayed template selection screen, it is determined whether data of a desired template selection number has been input (S603).
When it is detected that the data of the template selection number is input, the selection number is stored in the register n (S604), and the template data (n) in the data memory 104 is stored in the work RAM 103. Copy (S605).
Next, after displaying the initial rhythm creation screen, the present process is terminated (S606).
On the other hand, if the input of template selection data is not detected in the process of S603, it is determined whether or not the switch for canceling the process is turned on (S607).
Here, when the on / off state of the cancel switch is detected, this process is terminated. However, when the on state of the cancel switch is not detected, the process returns to S603 again.

<About waveform search processing>
Next, the detailed operation of the waveform search process (S503) in FIG. 5 will be described with reference to FIGS.
When the accompaniment tone color pattern to be a template is selected, it is next determined whether or not a switch for searching for a waveform is turned on (S701).
Here, if the ON of the waveform search switch is not detected, this processing is terminated, but if the ON of the waveform search switch is detected, a screen for searching for a waveform is displayed (S702).

Next, as a search target of waveform data to be sampled, whether to search from the waveform memory 105 in this apparatus or to search using the external database 110 via the communication circuit 109 is selected by a switch (S703). ).
Here, when the waveform search using the external database 110 is selected, the process proceeds to S801 in FIG. 8, but when the waveform search is selected from the waveform memory 105 in the apparatus. Then, it is determined whether or not the waveform data is stored in the waveform data area in the data memory 104 (S704).

When it is detected that the waveform data is stored in the waveform memory 105, the waveform data is copied and stored in the work RAM 103 (S705).
Then, the characteristics of the waveform data copied to the work RAM 103 are sequentially analyzed (S706).
On the other hand, if it is not detected that the waveform data is stored in the waveform memory 105, the process proceeds to S708.

Next, it is checked whether or not data that has not been subjected to waveform analysis still remains in the waveform memory 105 (S707). If there is unanalyzed data, the process returns to S705 to copy the waveform data to the work RAM 103. When all the characteristics of the waveform data are analyzed, a screen to end the waveform search is displayed (S708).
Next, it is determined whether or not the switch to return to the beginning of the waveform search process is turned on (S709).
Here, if the return switch is detected to be turned on, the process returns to S702 again. If the return switch is not detected to be turned on, it is next determined whether or not the switch for ending this process has been turned on ( S710).

  If it is not detected that the end switch is turned on, the process returns to step S709 again. However, if it is detected that the end switch is turned on, the display screen is returned to the initial screen for rhythm creation, and the process ends (S711).

Next, processing when the use of the external database 110 is selected as the waveform data search means in the processing of S703 will be described with reference to FIG.
First, the communication circuit 109 connects the present apparatus and the external database 110 via a communication line (S801).
Next, a file in the external database 110 is searched (S802), and it is determined whether or not the data in this file is waveform data. At this time, for example, the audio signal is converted into digital data in the form of AIFF (Audio Interchange File Format) whose file name extension is an audio file format, MP3 which is an audio compression method, or Windows (registered trademark) standard audio file. Whether it is WAV or the like, which is a storage format for recording the converted data, may be used as a criterion for determination (S803).

If it is not detected that the data in the file is waveform data, the process proceeds to step S806. If the data in the file is detected as waveform data, the waveform data in the file is communicated. The data is transferred to the work RAM 103 via the circuit 109 (S804).
Next, the characteristics of the transferred waveform data are analyzed (S805).
After this waveform analysis, it is determined whether there is an unsearched file in the external database 110 (S806).

  If it is detected that there is still an unsearched file in the external database 110, the process returns to S802 again to execute the file search. If it is not detected that there is an unsearched file, the process shown in FIG. Returning to the processing of S708, a screen for finishing the search of the waveform data is displayed.

<About waveform analysis processing>
Next, detailed operation of the waveform analysis processing (S706, S805) shown in FIGS. 7 and 8 will be described with reference to FIG.
First, extraction of the frequency characteristics of the waveform data stored in the work RAM 103 (S901), extraction of the envelope (S902), and extraction of the pitch (S903) are sequentially performed.
Next, “0” is set in the address register p of the standard tone color data (S904).

  The frequency characteristics, envelope, and pitch characteristics of the extracted waveform data are compared with the standard tone (p) frequency characteristics, envelope, and pitch characteristics stored in the data memory 104. Then, it is sequentially determined whether or not these waveform characteristics are similar (S905, S906, S907).

Each standard timbre in this standard timbre data group is based on the timbre of a specific type of musical instrument, and is set in consideration of the specifications of tone generation and tone generation functions by the sound source provided in the tone generator 106. This is characteristic data serving as a reference for the timbre waveform.
In the processing of this waveform analysis, the characteristics of the standard tone color are compared with the characteristics of the sample waveform data.
For example, in the comparison process of the frequency characteristics of the waveform data, whether or not the composition ratio of the harmonic component waveform included in the waveform data is similar to the composition of the harmonic component of the standard tone (p) waveform, or this waveform Whether the change point (attack, decay, sustain, release) of the amplitude position corresponding to the elapsed time of the data envelope is similar to the position of the envelope change point of the standard tone (p), or the pitch of the waveform data That is, it is compared whether or not the fundamental frequency indicating the pitch is within a predetermined frequency difference with respect to the fundamental frequency of the standard tone color (p) waveform.

When it is detected that these characteristics are similar to the characteristics stored in the standard timbre (p), a new empty area is secured in the data area of the search timbre (s) (S908). Then, the extracted waveform data in the work RAM 103 is stored in the data area of the reserved search tone color (s) (S909), and this process is terminated.
On the other hand, in the comparison process of S905, the comparison process of S906, and the comparison process of S907, the result of comparing each characteristic of the waveform data and each characteristic of the standard tone (p) is one of the frequency characteristic, the envelope, and the pitch. If it is detected that the characteristic is not similar, the process proceeds to S910 to advance the address register p of the standard tone color part (S910), and the value indicated by the address register p is the maximum number for storing the standard tone color. It is determined whether or not the value P has been reached (S911).

  Here, when it is detected that the value indicated by the address register p has reached the maximum value P of the address, this processing is terminated, but it is detected that the value indicated by the address register p has reached the maximum value P of the address. If not, the process returns to S905 to compare the waveform data with the characteristics of the standard timbre sequentially.

<About waveform assignment processing>
Next, the detailed operation of the waveform assignment process (S504) in FIG. 5 will be described with reference to FIGS.
First, when sample waveform data is searched and a standard tone color (p) having a characteristic similar to that of the waveform data is searched, whether or not a switch for assigning an accompaniment pattern corresponding to the waveform data is turned on. Is determined (S1001).

Here, when it is detected that the assignment switch is turned on, it is determined whether or not an accompaniment pattern template is selected (S1002).
On the other hand, if any template is selected, the head number “1” is set in the register n indicating the tone color pattern number to be the template (S1003).
Next, it is determined whether or not the timbre data corresponding to the selected template timbre pattern (n) exists in the data area of the search timbre (0) to the search timbre (S) (S1004).

  Here, when it is detected that the timbre data corresponding to the selected timbre pattern (n) is stored in the search timbre data group, the timbre pattern number address register n as a template is read. In step S1005, it is determined whether or not the value indicated by the address register n has reached the maximum number of template tone color patterns (S1006).

If the value indicated by the address register n does not reach the maximum number of template tone color patterns, the process returns to S1004 again to search for tone data corresponding to the tone color pattern (n) of the selected template. To go.
Then, when the value indicated by the address register n reaches the maximum number of template timbre patterns, a screen for assigning rhythm data corresponding to the timbre pattern (n) of the next selected template is displayed. The process proceeds to S1101 in FIG. 11 (S1007).

  On the other hand, if it is not detected that the switch to which the waveform selected in the process of S1001 is assigned is turned on, or if the timbre pattern as a template is not selected in the process of S1002, or is selected in the process of S1004. When it is detected that the template timbre pattern (n) does not exist in the data area of the search timbre group, in any case, “accompaniment pattern corresponding to waveform data can be assigned on the display screen”. After confirming the warning “Cannot do. Please try again from the beginning.” (S1008), the process is terminated.

Next, in the processing of S1101 in FIG. 11, the leading number “1” is set in the register n indicating the timbre pattern number as a template.
Then, one of the waveforms in the search timbre data group corresponding to the selected timbre pattern (n) is designated (S1102).
Next, the template tone color pattern address register n is incremented (S1103), and it is determined whether or not the value indicated by the register n has reached the value indicating the total number of template tone color patterns (S1104). .
If the value indicated by the register n has not reached the value indicating the total number of timbre patterns, the process returns to S1102 again, and the waveform of the search timbre data corresponding to the selected timbre pattern (n) is obtained. Specify sequentially.

When it is detected that the value indicated by the register n has reached the value indicating the total number of timbre patterns, a start switch for starting sound generation for auditioning the timbre pattern (n) of the selected template is turned on. It is determined whether or not (S1105).
When the start switch is turned on, "1" is set in the start flag STF and the rhythm data address register m, respectively (S1106).
Next, the timer T that counts the sound generation elapsed time for auditioning the timbre of the timbre pattern is started (S1107), and the leading number “1” of the timbre pattern is set in the address register n (S1108). The first event of the pattern (n) is read (S1109).

Then, it is determined whether or not the event of the read tone color pattern (n) is “OFF” (S1110).
Here, when it is detected that the event of the timbre pattern (n) is “off”, the CPU 101 proceeds to the processing of S1117 and advances the address register n, but the event of the timbre pattern (n) is “ If “off” is not detected, the waveform data is read from the data area of the search timbre waveform corresponding to the timbre pattern (n) (S 1115). The sound source is instructed to generate and generate a musical sound (S1116).

Next, the address register n is incremented (S1117), and it is determined whether or not the value indicated by the address register n has reached a value indicating the total number of timbre patterns stored in the data area (S1118).
If it is detected that the value indicated by the address register n has reached the value indicating the total number of timbre patterns stored in the data area, the process proceeds to S1201 in FIG. If it is not detected that the value has reached the value indicating the total number of timbre patterns, it returns to the processing of S1110 again to determine whether or not the event is “off”, and this event is “off”. If not, the waveform data of the next tone color pattern is read out, and the generation and generation of a musical tone for listening to the waveform data are instructed.

  On the other hand, if the start switch on is not detected in the process of S1105, it is detected whether or not the switch for canceling the process is turned on (S1111). If the stop switch is not detected to be on, the process proceeds to S1105. Returning to this, when it is detected that the stop switch is turned on, the start flag STF is set to “0” indicating the rhythm creation process (S1112), and the sound source is instructed to be silenced (S1113). ), The timer T is instructed to stop the count of the sound generation time for auditioning, and is cleared to zero, and this processing is terminated.

Next, when the value indicated by the address register n reaches a value indicating the total number of read timbre patterns, it is determined whether or not the start flag STF is “1” indicating reading of the timbre pattern (FIG. 12). S1201).
If the start flag STF does not indicate “1”, the process returns to S1105 in FIG. 11. However, if the start flag STF indicates “1”, the count time of the timer T is used for listening. It is determined whether or not the predetermined time α has elapsed (S1202).

If it is not detected that the count time of the timer T has passed the predetermined time α, the process returns to S1201. If it is detected that the count time of the timer T has passed the predetermined time α, then the timer T It is determined whether or not the count time of T has reached the pronunciation end time for trial listening (S1203).
When it is detected that the count time of the timer T has reached the sound generation end time, “1” indicating the head number is set in the address register n of the timbre pattern (S1204) and stored in the timbre pattern (n). The next event is read (S1205).
Then, it is determined whether or not the read event indicates a pronunciation “ON” (S1206).

  Here, if the event does not indicate “ON”, the CPU 101 instructs the sound source to mute the tone being generated (S1220), and then proceeds to the processing of S1209 to increment the address register n. However, if the event indicates “ON”, the timbre waveform data is read from the first data area of the search timbre data corresponding to the timbre pattern (n) (S1207), and is read out for the sound source. Based on the tone color waveform data, a musical tone is generated and an instruction is given to generate the musical tone (S1208).

Next, the address register n is incremented (S1209), and it is determined whether or not the value indicated by the address register n has reached a value indicating the total number of timbre patterns (S1210).
If it is detected that the value indicated by the address register n has reached the value indicating the total number of timbre patterns, the process returns to S1201 again, but the value indicated by the address register n indicates the total number of timbre patterns. If it is not detected that the value has been reached, the process returns to S1205 to read the event of the next timbre pattern (n).

On the other hand, when the count time of the timer T reaches the sound generation end time in the processing of S1203, the time count of the timer T is stopped and cleared (S1211).
Next, it is determined whether or not the “OK” switch for instructing the reading of the timbre pattern is turned on (S1212).
Here, when the “OK” switch is turned on, the process proceeds to S1301 in FIG. 13, and the head number “1” of the data area in which the timbre pattern is stored is set in the address register n.
However, if it is not detected that the “OK” switch is turned on, it is determined whether or not the switch for designating the next retrieval timbre data stored in the timbre pattern (n) is turned on (S1213).

If the next designated switch is not turned on, the process returns to S1105 in FIG. 11 to check whether the start switch is turned on. If the next designated switch is turned on, it is selected. After the head number “1” is set in the rhythm data address register m corresponding to the template data (S1214), another search timbre in the search timbre data group corresponding to the timbre m pattern of the rhythm data is set. Is designated (S1215).
Next, it is determined whether or not a separately designated tone color pattern is stored in the data area of the search tone color data (0) to the search tone color data (S) (S1216).

  Here, when it is detected that another timbre pattern designated in the search timbre data group is stored, the process returns to the processing of S1107 in FIG. Starts counting the pronunciation time. However, if it is detected that another timbre pattern designated in the search timbre data group is not stored, the rhythm data address register m is incremented (S1217), and the value indicated by the address register m is changed. It is determined whether or not a value indicating the total number of timbre patterns in the rhythm data has been reached (S1218).

  If the value indicated by the address register m does not reach the value indicating the total number of timbre patterns in the rhythm data, the process returns to S1215 to specify another search timbre among the search timbres corresponding to the timbre pattern m. However, when it is detected that the value indicated by the address register m has reached the value indicating the total number of timbre patterns in the rhythm data, a timbre is assigned to the rhythm pattern. The message “Please specify the tone pattern” is displayed for a certain period of time (S1219), and this process ends.

Next, in the process of S1301 in FIG. 13, the head number “1” is set in the address register n of the timbre data corresponding to the selected template data, and then copied to the timbre data area in the work RAM 103. The event of the template timbre pattern (n) being read is read (S1302), and it is determined whether or not the read event indicates “ON” (S1303).
If it is not detected that the read event indicates “on”, the process proceeds to S1305 to check whether there is a next event, but the read event may indicate “on”. When detected, the “start address” indicating the read start position of the search timbre waveform data corresponding to the timbre pattern (n) and the “end address” indicating the read end position are stored in the event area of the timbre data. (S1304).
Then, it is determined whether or not there is a next event in the timbre pattern (n) (S1305).

If it is detected that there is a next event, the next event is read out (S1306), and the process returns to S1303 to determine whether this event is “ON”. When there is no event, the template data address register n is incremented (S1307), and the value indicated by the address register n reaches the value indicating the total number of timbre patterns of the template data. It is determined whether or not (S1308).
Here, when it is detected that the value indicated by the address register n has reached the value indicating the total number of timbre patterns, an empty data area is secured in the rhythm data group (S1309), and the template in the work RAM 103 is reserved. After storing the tone color pattern data in the vacant area of this rhythm data group (S1310), this process is terminated.
On the other hand, if it is not detected in the process of S1308 that the value indicated by the address register n has reached the value indicating the total number of timbre patterns, the process returns to the process of S1302 and becomes a template stored in the work RAM 103. The event of the timbre pattern (n) is read out.

In this way, in the rhythm creation process, when the user selects a timbre pattern to be a model of accompaniment, the waveform data as a sample is read out from either the internal memory or the external database, By analyzing the characteristics of the waveform data, a standard tone color similar to the standard tone color is searched from the standard tone color group.
If it is found that the characteristic is similar to any standard tone color characteristic, the waveform data is added to the data area of the search tone color data group and stored.
With such a configuration, a wide variety of waveform data is accumulated in the search tone color data group, so that the types of tone colors that can be assigned to accompaniment patterns can be increased.
Furthermore, in this rhythm creation process, a predetermined tone is used for sequentially auditioning what kind of accompaniment musical tone each instrument tone of the template tone pattern desired by the user becomes, and any of the search tone waveform data for each tone pattern. If the musical tone of this timbre is satisfied, the timbre waveform data is assigned to the timbre pattern so that the timbre waveform data can be read out during reproduction of the rhythm accompaniment.

<About rhythm playback processing>
Next, the detailed operation of the rhythm reproduction process shown in S405 in FIG. 4 will be described with reference to FIGS.
First, an initial screen for reproducing rhythm data is displayed on the display screen of the display device 108 (S1401).
Next, it is determined whether or not the rhythm selection switch has been operated (S1402).

If the operation of the rhythm selection switch is not detected, the process proceeds to S1404. If the operation of the rhythm selection switch is detected, the rhythm selection number is set in the rhythm data address register M (S1403).
Next, it is determined whether or not the switch for starting reproduction of the selected rhythm pattern is turned on (S1404).

When it is detected that the start switch is turned on, "1" indicating rhythm reproduction processing is set in the start flag STF (S1405), then the timer T is cleared and the sound generation time is started (S1406).
Then, the head number “1” is set in the address register m of the timbre part of the rhythm data, and the first event stored in the timbre pattern (m) of the rhythm data (M) is read (S1408).
It is determined whether or not the read event is “OFF” indicating silence (S1409).

Here, if the read event does not indicate “OFF”, the CPU 101 designates the reading start position of the search tone waveform data based on the start address in the read event (S1412), and the waveform memory. The designated tone color waveform data is read out from the search tone color data group of 105, a tone is generated based on the rhythm of the rhythm data and the tone color waveform data to the sound source, and the tone generation of the tone is instructed ( S1413).
Next, after the rhythm data address register m is incremented (S1414), it is determined whether or not the value indicated by the address register m has reached the timbre pattern total number of rhythm data (S1415).

Here, when it is detected that the value indicated by the address register m has reached the value indicating the total number of timbre patterns of the rhythm data, it is determined whether or not the switch for stopping the rhythm reproduction process is turned on ( In S1501 in FIG. 15, if the value indicated by the address register m does not reach the value indicating the total number of timbre patterns of the rhythm data, the process returns to S1408 and the next of the timbre pattern (m) in the rhythm data (M). Read events.
On the other hand, if it is not detected in step S1404 that the rhythm playback start switch is turned on, it is next determined whether or not the mode switch is turned on (S1410).

  If it is detected that the mode switch is turned on, “0” indicating the rhythm creation process is set in the mode flag MODE, and then this process is terminated (S1411). If the mode switch is not turned on, the process proceeds to S1404. Returning to the process, it is checked whether the start switch is turned on again.

Next, if it is not detected in step S1501 that the stop switch for stopping the rhythm reproduction process is turned on, it is determined whether or not the timer T for counting the sound generation time indicates that a predetermined time α has elapsed (S1505).
If the timer T does not detect a value indicating the elapse of the predetermined time α, it is checked again in step S1501 whether the stop switch is turned on. When the indicated value is detected, it is determined whether or not the count time of the timer T has reached the time for instructing the end of sound generation (S1506).

Here, when the count time of the timer T reaches the time indicating the end of sounding, the process proceeds to the processing of S1502, and the sound is instructed. If the count time of the timer T has not reached the time for instructing the end of sounding, the rhythm data The leading number “1” is set in the address register m of the tone color pattern (S1507).
Then, the CPU 101 reads the next event in the timbre pattern (m) (S1508), and determines whether or not this event indicates the pronunciation “ON” (S1509).
When “ON” of the read event is detected, the reading start position of the timbre waveform data in the search timbre data group is designated by the start address in the event, and the timbre waveform data is read (S1510). After instructing the sound source to generate and generate a musical tone based on the read tone color waveform data (S1511), the process proceeds to S1512.

On the other hand, if it is detected that the read event is “off”, the sound source is instructed to mute the sound being generated (S1514), and the rhythm data tone pattern address register m is then walked. Advance (S1512).
Then, it is determined whether or not the value indicated by the address register m has reached a value indicating the total number of rhythm data tone color patterns (S1513).
Here, when it is detected that the value indicated by the address register m has reached the value indicating the total number of rhythm data tone color patterns, the process returns to S1501 to turn on the stop switch for instructing the end of this process. However, if it is not detected that the value indicated by the address register m has reached the value indicating the total number of rhythm data timbre patterns, the process returns to S1509 to search for “ON” of the event again. To do.

On the other hand, if it is detected in step S1501 that the stop switch for ending this process is turned on, an instruction is given to mute the musical tone being sounded (S1502), and the timer T that counts the sounding time is stopped and cleared. (S1503) After “0” indicating the rhythm creation process is set in the start flag STF, the process proceeds to S1404 in FIG.
In this way, in the rhythm playback process, when the user selects a desired rhythm after the end of the rhythm creation process and starts rhythm playback, an assignment is made for each timbre pattern of the template set in the rhythm creation process in advance. Based on the waveform data in the retrieved tone color data group, a musical tone is generated at the timing of the rhythm data.

  As described above, according to the present invention, the user designates a timbre pattern as a template for accompaniment, and the waveform of sound existing in nature such as wind and waves stored in a storage medium in the apparatus or an external database. Data, waveform data from human or animal utterances, waveform data of electronically generated musical sounds such as electronic musical instruments, waveform data such as waveform data of musical instruments of generally unknown natural instruments, etc. When this is specified, the characteristics of the waveform data are analyzed, and if this is similar to the characteristics of the standard tone color data, the waveform data as the sample is additionally stored as new search tone color data.

  By adopting such a configuration, the accompaniment for the accompaniment that becomes the desired template can be obtained even if the user is not familiar with accompaniment to the song, for example, based on a variety of timbre waveform data stored in the storage medium. Only by assigning desired timbre waveform data to the timbre pattern, an effect of creating and enjoying an accompaniment by an original rhythm pattern is achieved.

  In the embodiment of the present invention, the timbre waveform data of the search timbre data group is assigned to each of the template timbre patterns, and the rhythm performance is performed. However, the present invention is not limited to this, and the user specified in advance. The sample waveform data may be assigned as melody tone color waveform data, and the rhythm pattern tone may be generated based on the tone color waveform data assigned to the template tone color pattern in synchronization with the melody tone tone.

  In the embodiment of the present invention, the characteristics of the waveform data designated by the user are analyzed, and the search timbre group is obtained on condition that the characteristics are similar to all the characteristics of the frequency characteristics, envelope, and pitch of the standard timbre data. However, the present invention is not limited to this, and at least one of the characteristics of the standard timbre may be similar.

  Moreover, it can implement by programming the operation | movement procedure demonstrated in the said embodiment, and making a computer perform. At that time, the computer program can be supplied to the computer through a disk-type recording medium such as a hard disk, various memories such as a semiconductor memory or a card-type memory, or various program recording media such as a communication network.

The block diagram of the whole structure of this invention is shown. The data structure in the data memory shown by FIG. 1 is shown. (Part 1) The data structure in the data memory shown by FIG. 1 is shown. (Part 2) The basic process of this invention is shown. The subroutine of the rhythm creation process shown in FIG. 4 is shown. FIG. 6 shows a subroutine of the template selection process shown in FIG. FIG. 6 shows a subroutine of the waveform search process shown in FIG. (Part 1) FIG. 6 shows a subroutine of the waveform search process shown in FIG. (Part 2) The waveform analysis subroutine shown in FIGS. 7 and 8 is shown. FIG. 6 shows a subroutine of the waveform assignment process shown in FIG. (Part 1) FIG. 6 shows a subroutine of the waveform assignment process shown in FIG. (Part 2) FIG. 6 shows a subroutine of the waveform assignment process shown in FIG. (Part 3) FIG. 6 shows a subroutine of the waveform assignment process shown in FIG. (Part 4) The subroutine of the rhythm reproduction process shown in FIG. 4 is shown. (Part 1) The subroutine of the rhythm reproduction process shown in FIG. 4 is shown. (Part 2)

Explanation of symbols

101 CPU
102 Program ROM
103 Work RAM
104 Data Memory 105 Waveform Memory 106 Musical Sound Generation Device 107 Operator Group 108 Display Device 109 Communication Circuit 110 External Database

Claims (5)

Waveform data capturing means for capturing waveform data;
Pattern data storage means for storing timing pattern data of sound generation and mute of a specified tone color waveform;
Standard timbre data storage means for storing data representing the characteristics of the timbre specified by the timing pattern data stored in the pattern data storage means;
Determining means for determining whether the characteristics of the waveform data captured by the waveform data capturing means are similar to the characteristics represented by the data stored in the standard tone color storage means;
When it is determined by the determination means that the captured waveform data is similar, the accompaniment pattern is assigned by assigning the waveform data as the designated tone color waveform to the timing pattern stored in the pattern data storage means. Pattern generation means for generating;
An accompaniment pattern generation device characterized by comprising: The pattern data storage means stores a plurality of pattern data corresponding to different timbres, the standard timbre data storage means stores data representing the characteristics of the plurality of different timbres, and the pattern data storage means 2. An accompaniment pattern generation apparatus according to claim 1, further comprising selection means for selecting one pattern data from a plurality of stored pattern data and selecting data representing a timbre characteristic corresponding to the selected pattern data. .   2. The accompaniment pattern data generation apparatus according to claim 1, wherein the waveform data capturing means captures desired waveform data from a waveform data storage means in which a plurality of types of waveform data are stored.   2. The accompaniment pattern generation device according to claim 1, wherein the waveform data capturing means is connected to an external database for storing desired waveform data and captures the desired waveform data. A waveform data capture step for capturing waveform data;
A determination step for determining whether or not the characteristics of the captured waveform data are similar to the characteristics represented by the data stored in the standard tone color storage means;
When it is determined that the waveform data is similar to the designated timbre characteristic represented by the stored waveform data and the data stored in the standard timbre storage means, the pronunciation and mute timing of the specified timbre waveform stored in the pattern data storage means A pattern generation step of generating an accompaniment pattern by assigning the waveform data to the pattern as the specified timbre waveform;
Accompaniment pattern generation program for causing a computer to execute .

Images