JP2581370B2 - Automatic accompaniment device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a music apparatus, and more particularly to an automatic accompaniment apparatus for performing automatic accompaniment using a tonality analysis function for analyzing a tonality of a chord progression.

[0002]

2. Description of the Related Art Electronic musical instruments having an automatic accompaniment function are already known. In this type of electronic musical instrument, chord progression is input by sequentially designating chords as a combination of key codes (note numbers) from a performance input device such as a keyboard. Inside the electronic musical instrument, there is provided a chord constituent sound memory for storing a constituent sound of each chord of the chord set so as to indicate a pitch from a root (chord root note). By the chord root / type discrimination function, one of a plurality of input key codes (note numbers) is assumed to be the chord root, and the pitch from the root of the remaining key chords is obtained, and can be compared with the chord constituent sound memory. Generates specified chord component sound data. The chord root / type discrimination function searches the chord constituent sound memory for the generated designated chord constituent sound data. The type and route of the chord are identified by finding a match between the designated chord constituent sound data and the chord constituent sound data for a certain chord type in the chord constituent sound memory. In this way, a chord progression in which each chord is represented by a root and a type is obtained. Further, an accompaniment pattern memory for storing an accompaniment pattern is provided inside the electronic musical instrument. The accompaniment pattern includes horizontal (time) information and vertical (pitch) information of the accompaniment line. According to the accompaniment decoding function, the stored vertical information of the accompaniment pattern (vertical data of each accompaniment sound) is converted into data indicating a specific pitch (pitch) according to the identified chord type and route.

[0003]

In this type of apparatus , a code
When determining the accompaniment sound from the specified
Although the constituent sounds (pitch) itself are clear, this code
It is unclear what function the has. But
Therefore, the accompaniment sound is not necessarily
Can not select under the desired conditions. For example, for chord C Major (chord constituent sounds C, E, G)
Considering this , if this code has the function of I (tonic), the code C Majo as tonic
constituent sounds of the scale using r, ie, C, D, E,
It is desirable that F, G, A, and B be the conditions for selecting the accompaniment sound . If this code has the function of V (dominant) , use code C Major as the dominant.
That the scale of the component tones, Sunawa Chi C, D, E, F, G,
It is desirable that A, B ♭ be the conditions for accompaniment sound selection . This
If the code as it has the function of IV (subdominant), the code C Major as subdominant
The constituent sounds of the scale used, ie, C, D, E, F
It is desirable that ♯, G, A, and B be the conditions for accompaniment sound selection.
No. However, in conventional devices of this type,
Is unknown, so we can't determine what the code does,
The desired accompaniment sound could not be selected.

Accordingly, the applicant of the present application has filed Japanese Patent Application No. 3-68922.
In this issue, a tonality analyzer that analyzes the chord progression in real-time tonality and an automatic accompaniment device that performs automatic accompaniment using the analysis result (chord function type and key) of the tonality analyzer are proposed. According to this automatic accompaniment device, the above-mentioned problem can be solved and an accompaniment having a natural tonality can be provided. However, the accompaniment data forming means of the automatic accompaniment apparatus has a configuration in which, once a particular key is formed, an accompaniment pattern is formed, and then the accompaniment pattern is transposed by a key provided from the tonality analyzer. I have. The transposition referred to here (transpo
The processing of (se) is the keynote K of the specific key.
_REF (e.g. C) and the tonality given by the tonality analyzer
The difference ( _KANA - K) from the keynote _KANA (for example, G)
_REF ), in other words, the sound of K _ANA for K _REF
(For example, GC = 7 semitones)
Add to the pitch of each note of the formed accompaniment pattern (add
Process). Therefore, when the tonality analyzer detects the transposition from the chord progression, the range of the accompaniment after the transposition changes with respect to the range of the accompaniment before the transposition according to the difference in the key before and after the transposition (the difference of the keynote). However, there is a problem that sound skipping phenomenon occurs. Such skipping phenomenon, music intentionally its possible if the request is an exceptional manner the except for on preferably have na music. This problem also exists in other prior art automatic accompaniment devices having a tonality analysis function disclosed in Japanese Patent Application Nos. 62-333595 and 2-29787.

[0005]

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide an automatic accompaniment apparatus capable of maintaining a desired range of accompaniment before and after modulation.

[0006]

According to the present invention, a chord progression input means for inputting a chord progression, a tonality analysis of the input chord progression, and a modulation in the chord progression is detected by the tonality analysis. In an automatic accompaniment apparatus comprising a possible chord progression tonality analysis means and an accompaniment formation means for forming an accompaniment based on the analysis result of the chord progression tonality analysis means, the accompaniment formation means includes an accompaniment before and after the modulation. the pitch line, has a modulation compensation accompaniment means, for controlling a format that compensates for differences in the pitch formed between the tone before and after the modulation, the modulation compensation accompaniment means, tone glue
Accompaniment by key group that stores accompaniment pattern data
From the pattern storage means and the chord progression tonality means.
To the key group to which the key belongs
Record the accompaniment pattern data for each key group
Reading means for selecting and reading from memory means;
The accompaniment contents based on the accompaniment pattern data
Automatic accompaniment apparatus, comprising:
Is provided. According to this configuration, modulation (change of key)
The pitch line of the accompaniment before and after transposition is controlled in such a manner that the difference is compensated according to the magnitude of the difference between before and after the key, so that the above-mentioned sound skipping phenomenon can be avoided and the desired sound range can be maintained. Can be provided. Preferably, the generation means includes conversion data storage means for storing a table of pitch conversion data for converting pitch information by tonality, accompaniment pattern data from the reading means, and chord progression tonality analysis. receives the tonality information from the means, and the pitch converting data reading means for reading the voice pitch converting data corresponding from the conversion data storage means by using the pitch information and the tonality information included in the accompaniment pattern data,
Having. Such a configuration is effective in saving the amount of storage data required for forming the accompaniment. Further, according to the present invention,
A chord progression input means for inputting a chord progression;
Tonal analysis of tonal chord progression and tonality analysis
Chord progression tones that can detect modulation in chord progression
And analysis means, based on the analysis result of the chord progression tonality analysis means
Accompaniment forming means for forming an accompaniment based on the
In the playing apparatus, the accompaniment forming means may be used before and after modulation.
The accompaniment pitch line between the keys before and after transposition.
Modulation compensation controlled in a form that compensates for the difference in pitch formed
A compensation accompaniment means, and the modulation compensation accompaniment means has a reference accompaniment means.
Reference accompaniment pattern storage means for storing a performance pattern;
The pitch information contained in the standard accompaniment pattern
Keys included in the tonality information given by the advanced tonality analysis means
-Based on note type, chord type and chord function
The accompaniment pattern that is actually automatically played
Automatic accompaniment apparatus is provided, characterized in that it comprises a forming means for forming, a. In one configuration example, the forming means converts each pitch information included in the reference accompaniment pattern according to the type of keynote included in the tonality information provided from the chord progression tonality analysis means. A first conversion means for forming a first conversion accompaniment pattern, and converting each pitch information included in the first conversion accompaniment pattern into a chord type included in the tonality information provided from the chord progression tonality analysis means. A second conversion accompaniment pattern, that is, a second conversion means for forming an accompaniment pattern that is actually automatically played. This configuration also enables a desired accompaniment with a relatively small storage capacity. Preferably, the second conversion means accesses the pitch information of the first conversion accompaniment pattern as a first argument and a combination of a chord type and a function as a second argument, and sets a pitch from a key note or a chord root. Conversion table storing means for returning pitch data to be represented, and pitch conversion data from the conversion table storage means and data representing a pitch class of a key note or a chord root from the chord progression tonality analyzing means being arithmetically synthesized to perform a second conversion. Calculating means for forming pitch data of the accompaniment pattern.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Principle (FIG. 1A) FIG. 1A is a block diagram showing the principle of an automatic accompaniment apparatus according to the present invention. The chord progression input means 2 inputs a chord sequence in which each chord is represented by a root and a type. The tonality analysis means 4 analyzes the chord progression given from the chord progression input means 2. As a result of the tonality analysis, the function and key (key) of each chord in the input chord progression are obtained. Such tonality analysis means 4 is disclosed in Japanese Patent Application No. 3-68922.
The means 4 can detect modulation (key change) from chord progression as part of tonal analysis. For example, assume that CM → FM → GM → Am → DM → GM is given as a chord sequence from the chord progression input means 2. Tonality analysis means 4 determines the key code C from the first code CM to the fourth code Am for this code string and determines the code function / type F string as IM, IVM, VM,
It is determined as VIm. On the other hand, the fifth code D in the code sequence
M and the sixth code GM are determined that the key is G,
VM and IM are obtained as a column of the code function / type.
The modulation compensation accompaniment forming means 6 forms an accompaniment based on the analysis result of the tonality analyzing means 4. As a result of the tonality analysis of each chord, a chord function / type F (for example, IM) and a key KEY
(For example, C, G) is input to the modulation compensation accompaniment forming means 6. In the accompaniment forming means of Japanese Patent Application No. 3-68922 ,
6A is included in the pitch forming means for forming the actual pitch of the accompaniment sound.
The key KEY or information related to the key (a chord route related to the key via the chord function) is added to the output of the means 6A and the means for forming the pitch P (F) in the case of a specific key as shown in FIG. A transposition adding means 6B is included. In such a configuration, accompaniment output pitch sound depends on the key KEY
Modulation, which is a key change while music is progressing (MODUL
ATTION), that is, the accompaniment pitch line moves in parallel in response to a key change (modulation) occurring in the middle of the music . An example of the parallel movement is shown in E1. Key is C and code is C
It is assumed that the M-time accompaniment forming means forms an accompaniment having a CGEG pitch line. That is, the accompaniment pitch line is output from the means 6A for the key C and the code CM. Further, the value of KEY for the key C is set to 0.
Therefore, the output of the means 6A becomes the pitch line (CGEG) of the accompaniment output as it is. However, when the chord sequence described in the above example is given from the chord progression input means 2, the key of the last chord GM is determined to be G, so that the pitch line output (CGEG) of the means 6A is transposed and added. Result score G1 given G (value 7) to key input of means 6B
As shown in the right half of the figure, the accompaniment forming means pitch-lines the accompaniment output and outputs the GBDD. That is, the pitch line when the key is C and the code is CM (the pitch line before modulation) is transposed completely 5 degrees upward (only the difference between the key G and the key C ) when the key is G and the code is GM. (TR
ANSPOSE). It can be seen that the two pitch lines shown in the score E1 are skipped. The similar skipping is the adjacent code Am, DM
Also occurs between and. Such skipping phenomena of the accompaniment before and after the modulation are not musically preferable except for intentional exceptions. According to the present invention, the skipping phenomenon before and after modulation is caused by the difference in pitch formed between keys (keys) before and after modulation (in the above example, a perfect fifth formed between keys C and G). This problem can be solved by providing a compensating function 6C for controlling the pitch line of the accompaniment before and after the modulation in the form of compensating the above. An example of compensation is shown in score E2. In this case, the accompaniment pitch line CGEG for the key C and the code CM is BGDG for the key G and the code GM after modulation, and a smooth pitch flow is formed.

Aspects of the modulation compensation accompaniment forming means (FIG. 1B, FIG.
1C) The modulation compensation accompaniment forming means 6 described in FIG. 1A can be realized in several modes. In the first mode, a combination-based accompaniment pattern memory for storing an accompaniment pattern corresponding to a combination of a key and a chord function / type on a one-to-one basis is prepared. The accompaniment pattern for the specific combination is called out from the combination-based accompaniment pattern memory and reproduced. The key to this configuration is to provide the accompaniment pattern memory for each combination with accompaniment pattern data that compensates for the difference in pitch between keys.
However, this first mode requires a large storage capacity for accompaniment. FIGS. 1B and 1C show the second mode and the third mode of the modulation compensation accompaniment forming means 6, respectively. The modulation compensation accompaniment forming means 6M shown in FIG. 1B has a reference accompaniment pattern memory 60 for each key group. Memory 6
0 indicates a first-tone group reference accompaniment pattern memory 60-1 and a second-tone group reference accompaniment pattern memory 6 as shown.
0-2 ... Here, a key group is one having a specific key or two or more adjacent keys as elements. The selection module 62 selects a reference accompaniment pattern of a key group to which the key KEY belongs from the key group-specific memory 60 according to the key KEY given from the tonality analysis means 4. For example, if the first key group is a key group of B, C, and C #, when the key KEY is C, the first key group reference accompaniment pattern memory 60-1 is selected. Each tone type (pitch index) included in the reference accompaniment pattern selected via the selection module 62 is input to the conversion module 64. The conversion module 64 converts each of the input tone types according to the chord function / type F information given from the tonality analysis means, and generates a key KEY.
The pitch data expressed by the pitch from the (or chord root) is output. This pitch data is added to the key KEY (or chord root) by the adder 66 to become the actual pitch of the accompaniment. The key in this configuration is that a reference accompaniment pattern is prepared for each key group within a narrow range. The reference accompaniment pattern between the groups has pitch information that compensates for the pitch difference between the key groups, thereby avoiding the skipping of pitch lines before and after transposition.

The modulation compensation accompaniment forming means 6N shown in FIG. 1C is a reference accompaniment pattern memory 6 shared for all keys.
One. Reference accompaniment pattern comprises columns (pitch type columns) of pitch index as a sequence of pitch data. During the operation of the automatic accompaniment, the pitch index relating to the sounding is called from the reference accompaniment pattern memory 61 each time the accompaniment is sounded. This pitch index is input to the first conversion module 62. The first conversion module 63 converts the input pitch index according to the key KEY given from the tonality analysis means. That is, the first conversion module 6
Reference numeral 3 converts the input pitch index in a format that compensates for the difference between keys (keys), and outputs the converted index. The conversion index converted by the first conversion module 63 is input to the second conversion module 65. The second conversion module 65 converts (decodes) the conversion index in accordance with the information of the chord function / type F from the tonality analysis means, and generates a key KEY as pitch information suitable for the chord function / type F.
Outputs pitch data indicating the pitch from the (or chord root). This pitch data is added to the key KEY (or chord root) from the tonality analysis means in the adder 66. The addition result of the adder 66 indicates the accompaniment pitch actually played. In the third embodiment, the first conversion module 63 holds the key of modulation compensation. In terms of storage capacity, the first embodiment requires the largest storage capacity, and the third embodiment can form an accompaniment pitch line without skipping before and after modulation with the smallest storage capacity.

Hardware Configuration (FIG. 2) FIG. 2 shows a block diagram of a typical hardware configuration of an electronic musical instrument for realizing the automatic accompaniment apparatus of the embodiment. CPU10
0 executes the program stored in the ROM 102 to control the entire system. The ROM 102 stores fixed data (accompaniment pattern data, pitch change data, etc.) in addition to programs. The RAM 104 stores variables and temporary data and is used as a working memory of the CPU 100. The input device 106 includes a keyboard and other keys, switches, and volumes which are performance input devices of the present system (electronic musical instrument). The tone generator 108 electronically generates a tone signal under the control of the CPU 100. The sound system 110 receives a tone signal from the tone generator 108 and reproduces the sound. The timer 112 outputs a clock pulse to the CPU 100 at predetermined time intervals to enable the system to perform a fixed time process (for example, an automatic accompaniment process). The display device 114 displays system status, data, messages, and the like under the control of the CPU 100. Several embodiments will be described below, all of which are realized on an electronic musical instrument having a hardware configuration as shown in FIG.

First Embodiment (FIGS. 3 to 5) The automatic accompaniment apparatus according to the first embodiment includes an accompaniment pattern memory for each combination in the ROM 102 of FIG. 2, that is, a combination of a key (key) and a chord function / type. It has a combination-specific memory for storing accompaniment pattern data. FIG. 3 shows a part of the memory for each combination. That is, FIG. 3 shows the accompaniment pattern data of the C key, the D key, and the A key when the chord function / type is IM (each function is I and the type is M).
In practice, the combination-specific memory stores accompaniment patterns for all keys and all chord functions / types for each combination. Each word of each accompaniment pattern includes pitch data (for example, C4) and ON / OFF / bit data indicating whether to sound or mute. In the illustrated example, each accompaniment pattern is composed of 16 words (16 addresses). The 16-word accompaniment pattern is defined as a two-bar accompaniment pattern. Therefore, the music time per word is an eighth note. ON
The music time from the word including the bit to the word including the OFF bit determines the sound generation time of the sound.
FIG. 4 shows the operation of the first embodiment. The CPU 100 receives key press information for designating a chord from the keyboard (4-1).
As shown in 4-2, the type and route of the designated code are detected. Since the type of code and the technique of detecting the route are well known, the description is omitted. Subsequently, the CPU 100 executes a key note / function determination process as shown in 4-3. FIG.
An example of a key note / function determination process is shown in a flowchart in FIG. If the current key note is determined (5-1), the CP
U100 converts the new chord newly input from the keyboard into a function name (function / type) according to the current key note (5-2). Next, the CPU searches the same key note maintaining code table stored in the ROM 102 (5-3).
If the item corresponding to the function name generated in 5-2 is in the same key note maintaining code table, YES is satisfied in 5-4. This means that the key of the new chord is equal to the current keynote and 5
-The function generated in -2 is the correct function of the new code.
It means that it represents a type. If the retrieval of the same key note maintaining code table fails, the process proceeds from 5-4 to 5-5, and the CPU 100 changes the immediately preceding code to a function name according to the current key note. Then at 5-6 C
The PU 100 searches the parallel tone code sequence table stored in the ROM 102. In this sequence table, a set of code sequences from major to minor is written. If the parallel tone code sequence table has an item corresponding to the column of the function name generated in 5-2 and 5-5,
7 is established. Therefore, the key note and the function are determined, and the processing routine returns to the main routine. If the matching 5-7 is unsuccessful, the process proceeds to 5-8 to perform the pivot modulation inspection. RO in pivot modulation inspection
The pivot modulation code sequence table located in M102 is searched. The pivot modulation code sequence table stores a set of information of a code sequence representing a predetermined modulation. Basically, the CPU 100 regards the previous code as a pivot code, generates a function name of the previous code for some closely related keys, and pivot-modulates a column composed of each function name of the previous code and a function name of each closely related key of the new code. Search the code sequence table. If there is a corresponding item, the condition is satisfied in 5-9, but the condition is satisfied. Then, as shown in 5-10, the CPU transposes the current key note to a close key note (related key note). If 5-9 is not satisfied or the current key note is uncertain (5-1), 5-1
The process goes to step 1 to directly determine the key note and function of the new code with reference to the code type / function correspondence table stored in the ROM 102.

Returning to FIG. 4, key note / function determination processing 4
Subsequent to -3, the CPU 100 executes step 4-4.
That is, the CPU 100 stores the accompaniment pattern corresponding to the combination of the determined key (keynote) and the chord function name (function / type) in the accompaniment pattern memory (ROM 10
2). Subsequently, as shown in 4-5, the CPU 100 reads out the selected accompaniment pattern from the selected accompaniment pattern memory, and performs tone generation / silence processing to the musical sound generation device 108 according to the accompaniment pattern. Actually, the process 4-5 is a fixed time process, and the CPU 100 reads out the next element from the selected accompaniment pattern memory every time the music resolution time elapses, and if the element is a sounding event, the element (word) is used. Using the written pitch data, a pronunciation command is sent to the tone generation device 108, and if the element is a mute event, the pitch written in the element (the sound of the pitch being generated by the tone generation device 108) is output. A mute command is given to the tone generator 108 so as to mute. FIG. 6 shows the operation of this embodiment in comparison with the conventional one. CM from the keyboard as shown
When a code, ajor, AMajor, and CMajor are input, the first CMa is executed by keynote / function determination processing.
The section of ior is determined to be C-tone and INajor, the second section of AMajor is determined to be A-tone and IMajor, and the section of third CMajor is determined to be C-tone and IMajor. In other words, in this example, the modulation occurs from the C key to the A key and from the A key to the C key. In contrast to such tonality analysis results, in the conventional method, a pitch difference corresponding to a difference in tonality before and after transposition appears in accompaniment as shown in the figure. In the illustrated conventional accompaniment example, the pitch line of the accompaniment is CGEG in the first C-tone section, whereas the pitch line becomes AEC # E in the next A-tone section, and the pitch (tone range) of the C-tone accompaniment. In contrast, the A-tone accompaniment pitch (tone range) jumps. Further, when returning from the A key to the C key, a phenomenon that the pitch (tone range) jumps in the opposite direction occurs. Such a phenomenon of pitch (tone range) is not musically preferable except for intentional exceptions. On the other hand, according to the present invention, as shown in the score of the present embodiment in FIG.
The accompaniment pitch line of the next A-tone section becomes C # AEA with respect to the accompaniment pitch line of the tonal section CGEG, and the jump phenomenon of the musical range does not occur before and after the modulation, that is, before and after the modulation from the C-key to the A-key. , Smooth voice reading is obtained. This is ensured when returning from the A key to the C key. In the first embodiment, since the accompaniment pattern memory for each combination of key and chord function / type is prepared, not only the rhythm but also the shape of the pitch line can be set as desired for each combination. Therefore, for example, an accompaniment pitch line such as AC # AE can be generated in the A-tone section in place of the accompaniment pitch line in the illustrated A-tone section.

Modification of First Embodiment (FIGS. 7 and 8) However, the accompaniment pattern memory for each combination of key and chord function / type requires a large storage capacity. In a typical automatic accompaniment device, accompaniment of a plurality (for example, 20) of accompaniment styles is used as accompaniment music of a plurality of parts (for example, three parts) in a format depending on a chord in a chord progression, and a music section is an intro, a normal section, or a fill-in section. The accompaniment has different contents depending on whether it is the ending section or the ending section. When this is applied to the first embodiment, the total number of accompaniment patterns becomes the number of accompaniment styles (20) × the number of parts (3) × the number of chord functions / types (for example, 30) × the number of types of music sections (4). 720 for the numerical example shown in
0 types of accompaniment patterns are required. In the modified example described below, the accompaniment pattern data is shared as a reference accompaniment pattern commonly used for all chord functions / types,
A table called a pitch change table is used to change the reference accompaniment pattern to an accompaniment pattern suitable for each chord function / type. By using such a pitch change table, the storage capacity of the reference accompaniment pattern memory can be reduced to one thirty (30) of the number of chord functions / types as compared with a memory not using the pitch change table (memory for each combination). FIG. 7 shows the pitch change table in PCH []. The pitch change table PCH [] is a table that returns pitch data using the pitch index (tone type) from the reference accompaniment pattern memory as the first argument and the determined chord function / type as the second argument. This pitch data represents the pitch for the reference pitch (attached to the pitch index) in the reference accompaniment pattern memory in FIG. When this pitch data is added to the reference pitch data associated with the pitch index, desired pitch data (data representing the pitch actually accompanied) is obtained.

Therefore, the pitch data indicates the difference (difference pitch) between the target pitch and the reference pitch. The pitch index in the reference accompaniment pattern memory can be added, for example, next to C4 in the word (C4, ON) at address 0 in FIG. That is, C4 is the reference pitch in this case, and the data adjacent to C4 is the pitch index. (Note that pitch data such as C4 and G4 are not necessarily required in the reference accompaniment pattern memory. That is, these pitch data are omitted from the accompaniment pattern memory, and an appropriate reference pitch is set for each of the chord functions and types. A memory may be provided, and a desired pitch may be obtained by adding a reference pitch corresponding to the difference pitch read from the pitch change table, which leads to further saving of data.) FIG. 2 shows a flow of the accompaniment process in. 8-1 to 8-3 locate the next element of the selected accompaniment pattern. That is, j is incremented. (8-1) If j is the pattern size, (8-2)
j is returned to the value 0 indicating the first element of the selected accompaniment pattern (8-3). Next, the pitch data A included in the element (word) of the selected accompaniment pattern AM located in 8-4.
M _P Examine the [j] proceed to 8-5 if the pronunciation timing the sound of the word high index AM _T [j] with the key note, the pitch from the current code function type FDN has gained in the function determination processing The pointer i of the conversion table PCH [] is calculated. Next, at 28-6 ANT = AM _P [j]
The _{+ PCH [i] + TDN K} , seeking data ANT representing the actual sound high accompaniment. Actual accompaniment tone pitch data, i.e. in addition reference pitch pitch AM _P [j] to the pitch conversion adding the difference pitch from the table PCH keynote thereto, a pitch class TDN _K of the current keynote has gained in function determination processing I have ANT. Next, as shown in 8-7, the CPU sends a tone generation command including the pitch data ANT to the musical tone generator 108 so that the accompaniment tone of the pitch ANT is emitted by the musical tone generator 108.

Second Embodiment (FIGS. 9 to 14) Next, an automatic accompaniment apparatus according to a second embodiment will be described. FIG.
FIG. 9 shows a block diagram of a modulation compensation accompaniment function incorporated in the automatic accompaniment device of the second embodiment. In FIG. 9, a basic accompaniment pattern is stored in a basic accompaniment pattern memory REF. The basic accompaniment pattern is represented by a sequence of reference pitch data and a pitch type (pitch index) forming a pair of pitch sequences. The pitch change table PCT has substantially the same configuration as the pitch change table of FIG. 7, and outputs the difference pitch data ΔP with the pitch type as the first argument and the chord function / type as the second argument. A pitch decoding module that generates the actual pitch of the accompaniment using the basic accompaniment pattern memory REF, the pitch change table PCT, and information (key KEY, chord function / type) from the tonality analysis means is indicated by A. The pitch decoding module A can be composed of sub-modules A1 to A6. The processing order is A1, A2, A3, A4, A5,
The processing is performed in the order of A6. The first sub-module A1 converts the current pitch type read from the basic accompaniment pattern memory REF in accordance with the key KEY from the tonality analysis means to calculate the inverted pitch type. Next, the second sub-module A2 reads the calculated pitch of the turning pitch type (turning pitch) from the basic accompaniment pattern memory REF. That is, the submodule A2 searches the basic accompaniment pattern memory REF using the turning pitch type as a search key, and when the turning pitch type is found, extracts the turning pitch data attached to the turning pitch type. The extracted turning pitch data is used in a later sub-module A4. The third sub-module A3 reads the difference pitch data ΔP from the pitch change table PCT according to the turning pitch type calculated by the sub-module A1 and the current chord function / type from the tonality analysis means. That is, the submodule A3 calculates the address of the pitch change table memory PCT for storing the difference pitch data for the combination from the combination of the chord function / type and the turning pitch type, thereby accessing the pitch change table PCT to access the difference pitch data. Take out ΔP. The fourth sub-module A4 adds the difference pitch data to the turning pitch from the basic accompaniment pattern memory REF extracted by A2. Further, the fifth sub-module A5 adds a key to the result, and the last sub-module A6 adjusts the octave to form data representing the actual pitch of the accompaniment. The octave adjustment of the submodule A6 can be performed based on the reference pitch. That is, the pitch data (the pitch difference ΔP
And the current key KEY from the tonality analyzing means is added to the reference pitch, and if the difference is larger than a predetermined value (for example, a half octave), the difference is reduced by + or -1 octave. Only adjust the pitch data. A4
If the pitch data obtained through steps A5 and A5 is close to the reference pitch, it is output as it is as the actual pitch data of the accompaniment.
The pitch decoding module A shown in FIG.
02 is realized by the pitch decoding program provided in the CPU 02 and the CPU 100 which is a main body of executing the program.

FIG. 10 exemplifies, as KPM, a table of a calculation formula performed by the pitch decoding module A to obtain the actual accompaniment pitch. K1, k2, k in the pitch calculation table KPM
Reference numeral 3 denotes a current pitch type (PMN) read from the basic accompaniment pattern memory REF. C (0), C # shown in the row of Table KPM
(1), D (2), E ♭ (3), E (4), F (5),
F # (6), G (7), A ♭ (8), A (9), B ♭
(10) and B (11) are all tones KE that may be provided from the tonality analysis means (key note / function determination processing).
Y is represented. A formula for calculating the accompaniment actual pitch for the combination of the current pitch type and the key (for example, the current pitch type is k1) Items whose KEY is C are PNT + PCT (CC
M) (PMN) is written. PN in each formula
T represents the current reference pitch retrieved from the basic accompaniment pattern memory REF. PNT (Ki) represents the reference pitch of the pitch type Ki of the reference accompaniment pattern. Also P
CT (t) (y) represents differential pitch data for a PCT element for a chord function type x and a pitch type y, that is, for a combination of x and y. The CCM represents the current chord function / type given from the tonality analysis means (key note, function judgment processing). A direct method of realizing the pitch calculation table KPM is to provide a calculation routine for executing the calculation formula in the program of the ROM 102 for each calculation formula. The accompaniment process in that case is as shown in FIG. 14 described later.
FIG. 11 shows an example of the pitch change table PCT referred to in the calculation formula table KPM. FIG. 12 shows an example of the reference accompaniment pattern in a musical score. FIG. 13 shows FIG.
2 is used as a reference accompaniment pattern memory RE.
This shows an example of coding in F.

FIG. 14 shows a flowchart of the accompaniment process. This accompaniment process is executed every time the music resolution time elapses. First, at 14-1, the CPU 100 updates the current address (address for the current time) of the reference accompaniment pattern memory REF. Next, the CPU reads the word (current word) at the current address updated in step 14-2. Next, 14-3
The current word is examined, and if it is sounding timing, the current pitch type PMN included in the current word is obtained (14-4). More CPU
Is the current key KE obtained in the key note / function determination process
Y is taken out (14-5). Next, as shown in 14-6, the corresponding KPM routine (pitch calculation routine) is called and executed from PMN and KEY. As a result, pitch data PIT is obtained. Finally, the CPU sets the pitch data PIT
Is instructed to the tone generation device 108 by using (14-
7). As described above, according to the second embodiment, it is possible to avoid the skipping phenomenon of the accompaniment pitch line before and after the modulation while saving the storage capacity. The key to solving the sound skipping phenomenon is the sub-module A1 described with reference to FIG. That is, the sub-module A1 is a key KEY from the tonality analysis means.
The present pitch type from the basic accompaniment pattern REF is converted into an appropriate pitch type (inverted pitch type) so that the difference between them is compensated before and after the modulation. The description of the embodiment has been finished above, but various modifications and applications are possible within the scope of the present invention.

For example, in the above-described embodiment, the modulation sound skipping prevention means for preventing the pitch line of the accompaniment before and after the modulation from causing the sound skipping phenomenon has been described. You can also.
This is done, for example, by letting the user decide beforehand whether to select skipping or skipping, inputting the instruction from the input device, storing the content of the selection instruction in an appropriate flag, and storing the content of the automatic accompaniment. If no skip is selected by looking at the selection flag during the operation, an accompaniment is formed by the above-described method, and if skip is instructed, the current pitch read from the automatic accompaniment pattern memory REF is converted. The difference pitch data is taken out of the pitch change table PCT without using it as it is, the reference pitch of the current pitch type is added to the difference pitch data, and a key is added to the difference pitch data, thereby causing intentional skipping desired by the user. Can be done.

[0019]

As described above in detail, according to the present invention, the pitch line of the accompaniment before and after the modulation is controlled in such a manner as to compensate for the difference in pitch formed between the keys before and after the modulation. Therefore, it is possible to maintain the pitch range of the accompaniment pitch line regardless of any transposition, and to provide an accompaniment without skipping the pitch range.

[Brief description of the drawings]

FIG. 1A is a functional block diagram illustrating the principle of an automatic accompaniment device according to the present invention.

FIG. 1B is a block diagram showing one embodiment of a modulation compensation accompaniment forming means.

FIG. 1C is a block diagram showing another embodiment of the modulation compensation accompaniment forming means.

FIG. 2 is a block diagram showing a typical hardware configuration of an electronic musical instrument for realizing the automatic accompaniment device of the embodiment.

FIG. 3 is a diagram showing a part of data of an accompaniment pattern memory for each combination of key and chord function / type.

FIG. 4 is a flowchart showing the operation of the first embodiment.

FIG. 5 is a flowchart of key note / function determination processing.

FIG. 6 is a diagram showing an operation and an effect of the embodiment in comparison with a conventional technique using a musical score.

FIG. 7 is a diagram showing a pitch change table according to a code function / type.

FIG. 8 is a flowchart of an accompaniment process using a pitch change table.

FIG. 9 is a functional block diagram showing modulation compensation accompaniment forming means of the second embodiment.

FIG. 10 is a diagram showing a pitch calculation table according to the second embodiment.

FIG. 11 is a diagram illustrating a pitch change table used in the second embodiment.

FIG. 12 is a diagram showing an example of a reference accompaniment pattern in a musical score.

13 is a diagram showing an example in which the accompaniment pattern shown in FIG. 12 is coated on a reference accompaniment pattern memory.

FIG. 14 is a flowchart showing the operation of the second embodiment.

[Explanation of symbols]

2 Chord progression input means 4 Tonality analysis means 6 Modulation compensation accompaniment formation means 6M One mode of modulation compensation accompaniment formation means 6N One mode of modulation compensation accompaniment formation means 60 Accompaniment pattern memory by key group (Accompaniment pattern storage means by key group) 62 selection module (selection reading means) 64 conversion module (generation means) 61 reference accompaniment pattern memory (reference accompaniment pattern storage means) 63 first conversion module (first conversion means) 65 second conversion module (second conversion means) 66 Adder (arithmetic means) PCH [] Pitch change table (conversion table storage means) REF Reference accompaniment pattern memory PCT Pitch change table A Pitch decoding module A1 (first conversion means) A2 to A4, PCT (second conversion means) A5, A6 (Calculation means) KPM pitch calculation table

 ────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-1-173099 (JP, A) JP-A-2-29787 (JP, A) JP-A-5-61465 (JP, A)

Claims (5)

(57) [Claims] A chord progression input means for inputting a chord progression; a chord progression tonality analysis means capable of tonality analysis of an input chord progression and detecting modulation in the chord progression by tonality analysis; an automatic accompaniment apparatus comprising: a accompaniment forming means for forming an accompaniment based on the analysis result of the progressive tonality analysis means, and said accompaniment forming means, the accompaniment before and after modulation the pitch line, the tone before and after the modulation Modulation compensating accompaniment means for controlling in such a manner as to compensate for the difference in pitch formed therebetween, wherein said modulation compensating accompaniment means stores accompaniment pattern data for each key group.
In response to the key given from the chord progression tonality means, and
Accompaniment pattern data for the key group to which the key belongs
Select and read from the key group accompaniment pattern storage
Based on the read-out selection means and the read-out accompaniment pattern data
Generating means for generating an automatic accompaniment. 2. The automatic accompaniment device according to claim 1,
The generating means includes pitch conversion data for converting pitch information by tonality.
Conversion data storage means for storing a table of the above, accompaniment pattern data from the selective reading means,
Receiving the tonality information from the
Pitch information and the tonality information contained in the performance pattern data
The corresponding pitch conversion data from the conversion data storage means using
An automatic accompaniment device , comprising: a pitch conversion data reading means for reading data . 3. A chord progression input means for inputting a chord progression.
And tonality analysis of the input chord progression and tonality analysis
Chord progression to detect modulation in chord progression
Analysis means, and accompaniment based on the analysis result of the chord progression tonality analysis means.
An automatic accompaniment apparatus and a accompaniment forming means for forming a said accompaniment forming means, the pitch line of the accompaniment before and after modulation, before and after the modulation
To compensate for pitch differences formed between tones in
Modulation compensation accompaniment means for controlling includes a said modulation compensation accompaniment means, reference accompaniment pattern storage means for storing the reference accompaniment pattern
And the pitch information contained in the reference accompaniment pattern.
Included in the tonality information given by the
Key note types, chord types and chord functions
The accompaniment pattern that is converted based on the actual performance
Automatic accompaniment apparatus characterized by having a forming means for forming an. 4. The automatic accompaniment device according to claim 3,
The forming means converts each pitch information included in the reference accompaniment pattern into the code.
Included in the tonality information given by the
Converted according to the type of keynote, the first converted accompaniment pattern
First converting means for forming a tone pattern, and each pitch information included in the first converting accompaniment pattern is
Included in the tonality information provided by the chord progression tonality analysis means.
According to the combination of code type and function
The second conversion accompaniment pattern,
An automatic accompaniment device , comprising: a second conversion unit that forms an accompaniment pattern . 5. The automatic accompaniment device according to claim 4,
The second conversion means uses pitch information of the first conversion accompaniment pattern as a first argument,
A conversion table storage means which is accessed by using a combination of a chord type and a function as a second argument and returns pitch data representing a pitch from a key note or a chord root; a pitch data from the conversion table storage means; The keynote from the analysis means or data representing the pitch class of the chord route is arithmetically synthesized and the second
An automatic accompaniment apparatus, comprising: an arithmetic unit that forms pitch data of a conversion accompaniment pattern.