JPH087589B2 - Automatic code addition device - Google Patents

Description

Description: TECHNICAL FIELD OF THE INVENTION The present invention relates to an automatic chord adding device for automatically adding chord progression to a given melody.

[Background] Generally, multiple codes can be interpreted for a melody.

In view of this point, the applicant of the present application has proposed an automatic chord adding device having a function of discriminating chord candidates that can be added to a melody by applying appropriate music knowledge and analyzing the melody ( Patent application (1), April 6, 1988, name "Automatic code addition device"). More specifically, the device of the embodiment of the present application has, as the above-mentioned function, means for determining whether or not chord addition is possible by matching the chord constituent sound of the assumed chord and the melody with respect to each section (bar) of the melody. It has a chord addition possibility determination means based on the classification of chord sounds and a chord addition possibility determination means based on the scale, and the chords that pass the above three judgments are set as chord candidates that can be added to each section. Each of these chord candidates is a valid chord for each section in the meaning that can be interpreted by the musical knowledge incorporated in the chord discriminating means (melody analyzing means).

However, since a plurality of chord candidates can be extracted for one melody section, there is a problem in that the user is forced to select one chord candidate. If it is assumed that the number of melody sections is X and N chord candidates (in reality, the value of N depends on the melody of the section) are extracted per one section, chord progressions that can be added to the melody. The total number of combinations of (sequence of codes) is N to the Xth power, and when X = 16 and N = 2, it becomes a little over 60,000. Therefore, it becomes extremely difficult for a user with little musical knowledge to select one preferable chord progression among them.

Therefore, it is very preferable to add a function for reducing the chord progression selection work of the user to the above automatic chord adding device. In that case, it is desirable that this function has the ability to guarantee the naturalness of the chord progression to be selected. This is because the addable chord candidates extracted by the melody analyzing means are valid in terms of music knowledge when the melody section is used as a unit, but the chord candidates are relatively long because they are arbitrarily connected. This is because, if a chord progression for a melody is generated, an unnatural chord progression may be obtained as a whole. Such unnatural chord progressions are usually not considered chord progressions as desired by the user.

[Object of the Invention] Therefore, an object of the present invention is to form a sequence of discriminated chord candidates in an automatic chord adding device that may discriminate a plurality of chord candidates can be added for each section by analyzing a melody. To provide an automatic chord adding device capable of automatically selecting a natural chord progression from among a large number of chord progression combinations as a whole and reducing the chord progression selection work on the user side. Is.

[Summary of the Invention] In order to achieve the above object, the present invention provides an automatic chord adding device provided with a melody analyzing means for analyzing a given melody to determine chord candidates that can be added to each melody section. , Measure the distance of the tonality of each chord candidate determined to be able to be added to each melody section with respect to the melody tone, and select one chord among the chord candidates according to the measurement result. It is characterized by having done.

[Operation and Development of the Invention] According to claim 1, the distance measuring means measures the distance between the melody key and the tonality of each chord candidate, and the chord determining means selects one of the chord candidates in the melody section. 1
One code is selected based on the measurement result of the distance measuring means. Therefore, the chord progression generated (1 per interval
The information obtained by connecting one chord to another) is consistent in terms of tonality distance and has naturalness. Therefore, the user does not need to select one chord progression from among many combinations, and the intended purpose is achieved.

The second aspect relates to a preferred configuration example of the means for providing the melody key information referred to by the distance measuring means.
According to this, the tone of the melody is determined based on the entire distribution of the chord candidates extracted by the melody analysis means. For example, the tone of the melody is determined by the chord candidate having the highest frequency among the chord candidates extracted.

The provision of information about the key of the melody can be realized by other means, and the key of the melody may be directly input from the user, or may be independently determined without referring to the result of the melody analysis means. . However, in the case of the former, the user may input an incorrect melody key,
In the latter case, the processing of the key discrimination means becomes complicated.

Claim 3 relates to a more preferred configuration example for chord progression selection. According to this, a first parameter that limits the tonality distance (first tonality distance) between the melody key and the chord added to each section of the melody, and the tonality distance between the chords (second tonality). The tonality distance between the chord candidates before and after extracted by the melody analysis means (second
And a second distance measuring means for measuring the tonality distance of the. Then, the code determining means selects the first code from the code candidates.
Of the measurement results from the means for measuring the tonality distance and the means for measuring the second tonality distance, a code that yields a measurement result that meets the distance condition specified by the first parameter and the second parameter. The candidate is finally determined as the code to be added.

In this configuration, the first parameter contributes to the consistency of the overall chord progression, and the second parameter contributes to the naturalness of the connection when connecting the chords before and after. As a result, the chord progression defined by the determined sequence of chords is not only natural as a whole, but also locally.

Furthermore, in the case of this configuration, since the first parameter and the second parameter serve as extremely simple chord progress instruction input information for the user, the user can be freed from the troublesome work of selecting chords for each section. Since the chord progression generated changes depending on the setting of the first parameter and the second parameter, it becomes possible to provide a chord signal more suitable for the user.

According to claim 4, the distance measuring means measures the tonality distance between each chord candidate and the chord candidate that can be added to the preceding melody section, and the chord deciding means determines the chord candidate in the melody section. Among them, one code is selected based on the measurement result of the distance measuring means. Therefore, the generated chord progression (information in which one chord is successively connected per section) is consistent in terms of tonality distance and has naturalness. Therefore, the user does not need to select one chord progression from among many combinations, and the intended purpose is achieved.

[Embodiment] An embodiment of the present invention will be described below with reference to the drawings.

<Overall Configuration> First, the overall configuration of the automatic code adding apparatus according to the present embodiment will be described.
Shown in the figure. Reference numeral 1 is a CPU, which performs overall control for code addition. The work memory 2 is used to temporarily store code data and the like. The input device 3 is used for inputting melody data, designating a tonality distance, and the like. As the melody input device, for example, a keyboard (real time input) can be used. The monitor 4 is composed of a CRT, a tone forming circuit, a sound system, etc., and displays and outputs melody and chord (accompaniment). The melody data input from the input device 3 is stored in the melody data memory 5. In the production rule data memory 6, data expressing musical knowledge for classifying non-harmonic sounds included in a melody (production rule when a chord candidate and melody characteristics (pitch progression, position of harmonic sound, etc.) are given. ) Is stored. The chord component sound memory 7 stores chord component sound data of each chord type.

The automatic chord adding device of this embodiment has a function of analyzing a melody for each section (bar) of the melody and extracting a code that can be added to the melody of each section. Specifically, this feature has knowledge (expressed in production rules) to classify non-harmonic sounds,
In order to check whether each chord can be added to each section of the melody, the melody is divided into chord sounds (chord constituent tones) and chord constituent tones in order to check whether each chord is added or not. In order to analyze the character (type of non-harmonic sound), the situation of the melody before and after each chord-composing sound is calculated by a function, and the non-harmonic classification knowledge is applied to the situation. When the hypothesized chord is correct (when it is a chord that can be added), the characteristics of all external sounds constituting the chord can be specified, but when it is incorrect, unclassifiable sounds (illegal sounds) are detected. Therefore, when a sound that cannot be classified is detected, it is possible to exclude a code that cannot be added by making an error in the assumption of that code.

The number of impossible code candidates obtained in this way is 1
There can be more than one for each interval. Therefore, if this information is used as the final result of the automatic chord adding device, the user has to select chords for each section in order to obtain chord progression, which makes the work difficult.

Therefore, in the present embodiment, information input by a simple operation from the user is received, and 1 section per section is accordingly received.
A function to automatically select one code from code candidates is added. In detail, as the above input information,
From the input device 3, the instruction information (RG described later) for the tonality distance between the melody key and the chord of each section and the instruction information (CG described later) for the tonality distance between the preceding and following chords are input. On the other hand, in the CPU 1 of the embodiment, the key corresponding to the most frequent code candidate among the chord candidates extracted for each section is determined as the melody key, and the tonality distance between this key and the chord candidate of each section is determined. In addition to the measurement, the tonality distance between the code candidates for the front and back sections is measured. Then, the measurement value of the tonality distance that best matches the tonality distance designated by the input device 3 is searched for, and the code candidate that brings the measurement value is selected as the code of the final result.

<Overall Operation> FIG. 2 shows a general flow of this embodiment. First, 2
At -1, the CPU 1 reads the melody data in the melody data memory 5 and the production rule data in the production rule data memory 6 into the work memory 2 (note that the instruction information RG and CG regarding the tonality distance from the input device 3 has already been read. Be assumed). Next 2-2
The non-harmonic sound analysis is performed for each measure, and the usable chord candidates are determined. In 2-3, the appearance frequency of each code candidate determined to be usable is calculated for each code. Then, in 2-4, the chord candidate having the maximum frequency among the appearance frequencies is searched for to determine the key of the melody. Finally, in 2-5, measure the measurement distance between the chord and the chord candidate of each measure, and the measurement distance between the chord candidates before and after, and use the measured values for the specified tonality distance parameters RG and CG. The chord progression is generated by finding the best matching measurement value, finding the chord candidate given the measurement value one by one.

<Data Format> FIG. 3 shows a format example of melody data and chord data used in the following flow. In this example, the melody data is represented by a string of pitch data and pitch length data, and the pitch data is 16 bits long, the upper 8 bits represent an octave, and the lower 8 bits are C for each semitone. Expressed by an integer value that is incremented by 1, the shortest note length is set to "1" and doubled note length is "2".
It is represented by an integer value such as.

On the other hand, the code data is 16 bits long, and the upper 8 bits represent the type (code type) and the lower 8 bits represent the root note of the chord.

<Reading of Melody Data> FIG. 5 shows the details of reading the melody data executed in 2-1 of the general flow of FIG. FIG. 4 shows an example of melody data (a song of Beethoven's joy) stored in the melody data memory 5. As a result of the processing in FIG. 5, the i-th pitch data is stored in MDi, the i-th pitch data is stored in MRi, and the total number of melody sounds is stored in MDNO.

<Reading Production Rule Data> Details of reading the production rule data executed in 2-1 of the general flow are shown in FIG. The production rule data memory 6 is shown in FIG.
An example of production rule data stored in FIG.

The entire production rule is based on music knowledge for classifying non-harmonic sounds included in a melody (Fig. 6 (b),
(See (c)), and each production rule data is the lower limit data Li as the data that defines the preamble of the rule, and the function instruction data that indicates the type of function.
Xi, upper limit data Ui, and data as the conclusion part of the rule
It has Yi and Ni. The function is a numerical representation of the characteristics of the melody to be analyzed, an example of which is shown in FIG. 14 described later.
It is a premise (proposition) of the production rule that the value Fxi of the function represented by the data Xi is equal to or more than Li and equal to or less than Ui (Li ≦ Fxi ≦ Ui), and the conclusion when this premise is satisfied is data It is indicated by Yi, and the conclusion when this premise is not satisfied is indicated by data Ni. When the data Yi or Ni has a positive value, that value indicates the number of the production rule to be referred to next in the forward inference described later, and when it has a negative value, it is not summed by its absolute value. The type of voice sound is expressed. Forward inference always starts with one rule, and this rule is called a rule. Forward finding ends when it finds a negative conclusion Yi or Ni. This negative value includes data representing an illegal sound (sound that does not correspond to any non-harmonic sound) indicated by ill in FIG. 6 (a). As will be described later, when such an illegal sound comes to the conclusion of inference, it is considered that the assumption of the chord, which is the premise of inference, is incorrect, and it is determined that the chord cannot be added to the rody. .

In the case of the address assignment of the production rule data shown in FIG. 6, the data of each production rule is stored in five consecutive addresses starting from the address of the lower limit data Li. For details, Li is an address divisible by 5.
However, the data of Xi is stored in the address of the remainder 2 divided by 5, Ui is stored in the address of the remainder 2, Yi is stored in the address of the remainder 3, and Ni is stored in the address of the remainder 4.

In FIG. 7, ruleno is set to the total number of production rules.

<Code Judgment> FIG. 8 shows a flow of code judgment executed in 2-2 of the general flow of FIG.

First, in 25-1, the bar number BAR of the measure is expected to be "1", and 25-
In step 2, the start position Ps and end position Pe of the melody data of the bar indicated by BAR are obtained. In 25-3, the root ROOT of the chord is expected, and in 25-4, the chord type TYPE is expected. And
In 25-5, the non-harmonic sound analysis is performed on the melody data from the start position Ps to the end position Pe of the bar assuming the chords defined by ROOT and TYPE. 25 results
If it is checked in -6 and the analysis result does not include any sound indicating unclassifiable (illegal sound), it is OK (addable). Only when OK, the root ROOT data ROOT and the type data TYPE at that time are pushed to the stack CSTK (25-7, 25-8). Therefore, the appendable code is not recorded in the stack CSTK. This process is repeated for all the chords for the bar indicated by BAR (25-9, 25-10, 25-11, 25-12). When all the chords that can be added to the one-measure melody have been extracted, an identification mark (separation mark) representing the one-measure break is pushed onto the stack CSTK (25-13). Then, the above processing is performed for all measures (25-14, 25-1).
5) When all measures are completed, push the end mark onto the stack (25-16).

As a result, all codes that can be added to the first measure are sequentially stored in the stack CSTK in pairs of ROOT and TYPE starting from the beginning, and the code data that can be added to the second measure across the delimiter mark. , And so on, followed by the end mark after the last delimiter mark.

<Selection of Melody Bar> Details of selection of the bar position of the melody executed in 25-2 of FIG. 8 are shown in FIGS. 9 to 11. With this process,
In Ps, the number of the melody data {MDi}, {MRi} of the first note of the bar of interest is entered,
Pe contains the number of the melody data that is one note before the first note of the measure next to the one of interest. The information of Ps and Pe defines the range of the melody data to be investigated in the non-harmonic sound analysis processing 25-5 described below.

<Nonharmonic analysis> Fig. 12 shows the details of the nonharmonic analysis performed in 25-5 of Fig. 8.

First, 33-2 initializes the non-harmonic counter j,
At -3, the note counter i of the melody data is set to the head position Ps of the bar of interest. 33-4, paying attention
Determine if the th note is a chord. Harmonic sounds are the same as the constituent sounds of the chord candidate. If it is not a harmony tone, the non-harmonic tone counter is incremented (33-5),
The function F representing the characteristic element of the melody centering on the note of interest is calculated (33-6), the forward inference by the production rule is executed (33-7), and the conclusion about the note of interest is set to RSj. Substitute (33-8). Note counter i
Is incremented (33-9), and the process from 33-4 onward is repeated until the last note Pe of the measure of interest is reached (33-10).

Details of the process 33-4 are shown in FIG. In 34-1, the chord data is converted into chord component sound data. This conversion is performed by using the type data TYPE of the chord data as a pointer, and from the chord component tone memory 7 to the component tone data expressed in the C tone (for example, 00001001 for major).
0001) and read it as root data RO of chord data.
It is performed by turning to the left by the number indicated by OT. In 34-2, the melody pitch data MDi is converted into data having the same format as the constituent sound data CC generated in 34-1. CCΛmm
When the result of (logical product) is not 0, the same sound as the melody sound is included in the chord constituent sounds. That is, MDi is a harmony sound (34-3, 34-5). MD when CC Λ mm = 0
i is a non-harmonic sound (34-3, 34-4).

Details of calculation 33-6 of F in FIG. 12 are shown in FIG. In this example, the function F is as follows: F ₁ : how many notes (non-harmonic) the note of interest is before (harmonic position of the rear) F ₂ : how many notes before the note of interest or harmonic sound is located (position of the front harmonic tones) F _3: number F of nonharmonic tones in the front harmonic tones to the rear harmonic tone _4: pitch difference between the front harmonic sound and rear harmonic tones F _5: forward harmonic sound and rear harmonic tone height of nonharmonic tones distribution between the F _6: forward harmonic tone from whether high melody sound to the rear harmony sound is changed monotonically F _7: and the rear harmony sound Pitch difference with the preceding note (pitch progression to the rear chord) F ₈ : Calculate the pitch difference with the other person from the front chord (pitch progression from the front chord) There is. Besides this, information indicating weak or strong beats,
Information that distinguishes note lengths may be added to the set of functions F.
Flowchart of calculation of each function F (Figs. 15 to 22)
(Fig.) Is clear from the description of itself, and therefore its explanation is omitted.

The details of the forward reasoning in 33-7 of FIG. 12 are shown in FIG. First, the rule number pointer P is set to "1" which indicates the root rule among the production rules (44-1). Then, it is checked whether or not the premise part (Lp ≦ Fxp ≦ Up) of the rule indicated by the rule number pointer P is satisfied. If not, the data Np of the negative conclusion part of the rule is used as a pointer to the next rule.
However, when the data Yp and Np are negative and values, the final conclusion has been reached, so the absolute value (−Yp, −Np) is used as the conclusion register as an identifier of the type of dissonant sound or an identifier of illegal sound. Set to. According to the flow, the condition Lp of 44-3>
When the condition Fxp> Up of Fxp or 44-5 is satisfied, the premise part Lp ≦ Fxp ≦ Up of the rule P is not satisfied, so the data Np of the negative conclusion part of the rule is substituted into a (44-4, 44 -6),
In other cases, since the premise part is established, the data Yp of the affirmative conclusion part of the rule P is entered in a (44-2). Substituting this a into P (44-7), when P is a positive value, it returns to the inspection of the next rule as a rule number pointer, and when P is negative, -P is classified as a non-harmonic sound. The result (44-8, 44
-9).

<Frequency Counting and Key Determination> FIG. 24 shows details of the frequency counting process of the code candidates that can be added, which is performed in 2-3 of the general flow (FIG. 2). This processing is a pre-processing for key determination performed in 2-4 (in the present embodiment, the chord candidate having the highest frequency among the chord candidates that can be added in the processing 2-2 is set as the melody key. Has been decided).

As described above, the TYPE data follows the ROOT data on the stack CSTK by the processing of 2-3, and the code is defined by this pair, and the ROOT data (the code that can be added remains in one bar remains after the TYPE data. If there is a next measure, the ROOT data (if there is a next measure, the root note of the chord that can be added to the next measure)
Or end mark continues. Therefore, the root note d ₁ ,
Prepare variables Dd ₁ and d ₂ for counting the code of tie d ₂ , scan the stack CSTK from the beginning to the end mark, and in the meanwhile, correspond to each pair reading of ROOT data and TYPE data. By incrementing Dd ₁ and d ₂ , the frequency of code candidates can be obtained for each code. Figure 24 shows the flow according to this.
When it becomes NO, the stack data d ₁ read in 47-2 is RO
This is for OT data. TYPE data always follows ROOT data, so read data d ₂ (47-5)
The code counters Dd ₁ and d ₂ indicated by the data d ₁ are incremented (47-6). When the end mark is read, the processing is complete (47-3).

FIG. 25 shows a flow for finding the chord having the maximum frequency on the frequency table {D _ROOT , _TYPE } obtained as described above and determining the key. Details of the processing 2-4 of the general flow Is. Initialize max at 48-1, 48-2, 48-3, 48-8, 48-9, 48-10, 48-11
And all the root sounds, as you can see from its related loops
Scanning ROOT, type TYPE, and in the middle of scanning, the frequency count D _ROOT , _TYPE of the code of the current _ROOT , _TYPE ,
If the value is higher than the maximum frequency max until then (4
8-4), while updating max with that value (48-5),
The ROOT and TYPE data at that time are substituted into R and T, respectively. (48-5, 48-7). Therefore, as a result of the processing of FIG. 25, the root note and type information of the chord with the highest frequency are set in R and T, respectively, and the key of the melody is determined by this. For example, R is the note name of C, and T
Indicates a major type, C major is the melody key, and R is C and T is minor, the C minor is the melody key.

<Chord Progression Decision> FIG. 26 shows details of chord progression decision processing executed in 2-5 of the general flow.

From the beginning to the stack CSTK at the stage of entering the flow shown in the figure
All chord candidates that are judged to be able to be added to each bar in order of bars with a bar line delimiter between the end mark addresses are root root and type TYPE.
Are arranged in the form of two data pairs. Further, the root note and type data of the chord (key decision code) corresponding to the key are set in R and T. Further, a parameter for limiting the tonality distance between the key input from the input device 3 and the chord of each measure and a parameter for limiting the tonality distance between adjacent chords are set in RG and CG, respectively. The flow shown in the figure generates one chord progression {CD _BAR } based on these pieces of information.

In FIG. 26, min is the minimum value when the following calculation || dist1 | −RG | + || dist2 | −CG | is performed for each chord candidate that can be added to the melody of the bar. This is a variable to be set and is initialized each time a new measure is checked. In the above equation, dist1 is a measurement value of the adjustment distance from the key (R, T) of the code candidate, and dist2 is a measurement value of the tonality distance from the previous code of the current code candidate. The root note d ₁ and the type d ₂ of the chord candidate when the minimum value is given are stored in ROOT and TYPE, respectively, and are transferred to the variable CD _BAR for storing the chord for the bar number at that time.

To add to the explanation, when going to 49-7, the data of stack CSTK (in d ₁ ) read in 49-4 is neither the end mark nor the delimiter mark.
Considering the storage order of the above data, the data d ₁ is root data. Therefore, at 49-7, the data of the type at the next address is read at d ₂ . Therefore, (d ₁ , d ₂ ) indicates a code candidate. Therefore, in the next 49-8, the tonality distance between the key (R, T) and this code candidate (d ₁ , d ₂ ) is measured by the method described later and set in dist1, and 4
In 9-9, the tonality distance between the definite chord of the previous bar and the chord candidate (d ₁ , d ₂ ) of the current bar is measured in the same manner and set to dist2, and in the above 40-10, || dist1 |- RG | + || dist2 | -CG | is calculated and assigned to a. Then, the value of a and the value of min are compared in 49-11, and if the direction of a this time is small, mi
Substitute a for n, and substitute root note d ₁ and type d ₂ at that time for ROOT and TYPE, respectively (49-12 to 49-14).

The data on the stack CSTK contains all the code candidate data that can be added to one bar between the delimiter marks. Therefore, when the delimiter mark is detected in 49-6, among the code candidates that can be added to the measure indicated by BAR in ROOT and TYPE, the code candidate that minimizes the value of the above expression. That is, it contains the root note and type data of the chord candidate that best fits the tonality distance between the chord and the chord specified by the user and the instruction condition (RG, CG) regarding the tonality distance from the previous chord. . This is the chord decided for the melody of bar number Bar. So, proceed to 49-15, re-initialize min to check the next measure, and measure the bar number Bar at 49-16.
To move the ROOT and TYPE data to OLDROOT and OLDTYPE as the root note and type of the previous chord (49-
17, 49-18), and at 49-19 ROOT and TYPE
The code data is obtained from and is assigned to the array element CD _BAR-1 .

As a result, the array {CD
_BAR } contains information indicating one chord progression for the entire melody.

<Measurement of tonality distance> Figure 27 shows the details of the tonality distance measurement performed at steps 49-8 and 49-9 in the chord progression determination flow (Figure 26). 28th
The figure shows the definition of the tonality distance between the chords used in the flow of FIG. As can be seen from this figure, the tonality distance between the chords in parallel relation (for example, Am and c) is zero. Further, the tonality distance from the reference of the code completely 5 degrees above the reference is -2, and the tonality distance of the code 5 degrees below the reference is +2.

In 50-1 of FIG. 27, the root note data of the reference chord is entered in st. The standard code here is 49-8 (26th
In the case of (Fig.), It is the key code (R, T), and in the case of 49-10, the code of the previous measure (OLDROOT, OLDTY).
PE). In 50-1, the root note data of the chord measured in a ₁ and a ₂ , that is, the root note data d1 of the chord to be measured in 49-8 and 49-9 is set. As shown in 50-5, the root data of a ₁ is 50-3 to 50
Each time the loop around -6, is turn on 5 degrees, a ₂ root data is turning under 5 degrees (around the ring shown in FIG. 28 counterclockwise, or corresponding to clockwise) . 50
In -3, a ₁ = st is satisfied when the root note of the chord to be measured is turned up five times i times, and in 50-4, a ₂ = st is satisfied. The root sound of i
It is when it turns to the bottom of the reference chord when turning down 5 times. Therefore, for the former, i × 2 is added to x as the adjustment distance (50-7), and for the latter, i × 2 is added to x.
Insert (-2). Up to this point, what is obtained in x is
It is the tonality distance of the measurement code when the reference code and the code to be measured are both major and minor. Therefore, in the processing from 50-9 to 50-17, distance correction is performed when the systems of both codes are different. For example, the standard code is Am and the code to be measured is Gmaj
If so, x = + 4 due to 50-7. This must be, for example, x = -2 in the definition of the tonality distance in FIG. In this case, in FIG. 28, 50-10
From 50-11 to 50-13, x = -2 by x = x-6
Is obtained. Also, assuming that the reference code is Cmaj and the code to be measured is Bmin, x = -10 is obtained by 50-8. According to FIG. 28, x = -4 must be obtained. In this case, in FIG. 50, 50-14 to 50-
Proceeding to 15, 50-17, x = x + 6 gives x = -4.

<Modifications> The description of the embodiments is completed above, but the present invention is not limited to the above embodiments, and various modifications and changes can be made.

For example, as a melody analysis, in the above embodiment, it is determined whether or not a code can be added to the melody of each section based on the classification knowledge of the non-harmonic sounds. However, in combination with this, or in place of this, it is assumed from the chord. It is possible to use a means for determining whether or not a chord can be added based on the relationship between the available note scale and the melody constituent sounds in each section, a means for matching the chord constituent sounds with the section melody constituent sounds, and the like. In short, the present invention is applied to an arbitrary melody analysis unit having a possibility of extracting a plurality of chord candidates from one section as a chord that can be added to a melody as a result of determining whether or not a chord can be added to each section. be able to.

Further, in the above embodiment, in order to automatically select one chord from all chord candidates determined to be able to be added to the melody of the section, the melody key and the adjustment distance between each chord candidate are set. Measure the tonality distance between adjacent chord candidates before and after, and select the chord candidate that gives the measurement value that best fits the information RG and CG specified by the user from the measured values of these adjustment distances. There is. This is a preferable configuration in order to easily provide the user with a desired chord progression, but the above user specified information is not necessarily required from the viewpoint of the present invention. For example, it is predetermined as RG and CG. You may use the value. Further, as the tonality distance, the tonality distances of the front and rear chords may not be considered. Regarding the tone of the melody, the user may directly input the tone of the melody if the musical score of the melody is available, or without referring to the information of the chord candidates that can be added to the melody of each section. A means for discriminating the key may be used. As another example of the key discrimination means based on the information of the chord candidates that can be added to the melody, assuming each chord candidate as a reference, the adjustment distance of the remaining chord candidates is measured, and the number of chord candidates on the plus side is calculated. The reference chord when the number of chord candidates on the negative side is most balanced may be determined as the melody key.

[Effect of the Invention] According to claim 1, in the automatic chord adding device provided with the melody analyzing means for analyzing the given melody and discriminating chord candidates that can be added to each melody section, it is determined that the chord can be added. The chord progression is obtained by measuring the tonality distance between each chord candidate and the key of the melody and selecting one chord for each melody section from each chord candidate based on the measurement result. Therefore, there is an advantage that the user can be provided with a natural chord progression whose tonality is controlled, and the chord progression selection work performed by the user can be reduced.

According to the second aspect, since the means for discriminating the key is used as the key discrimination means based on the distribution of all the chord candidates which the melody analyzing means has determined to be addable, in addition to the effect of the first aspect, the discrimination of the key is performed. Can be reliably obtained with a simple configuration.

According to the third aspect, the parameter for limiting the tonality distance between the chord and the melody key and the parameter for limiting the tonality distance between the preceding and following chords can be input from the input means, and each chord candidate and melody of the section can be input. In addition to the means for measuring the tonality distance to the key, a means for measuring the tonality distance between the preceding and following chord candidates is provided, and the chord deciding means designates both parameters input from these measured values. By selecting chord candidates that correspond to the measured values that meet the conditions, it provides not only the overall tonal distance but also the tonal distance between chords with controlled and natural chord progression. be able to. Furthermore, since the generated chord progression can be changed by an extremely simple input from the user, it is possible to efficiently provide the chord progression that the user desires.

According to claim 4, in the automatic chord adding device provided with a melody analyzing means for analyzing a given melody and discriminating chord candidates that can be added to each melody section, each chord candidate determined to be addable Measures the tonality distance with chord candidates that can be added to the previous melody section, and selects one chord for each melody section from each chord candidate based on the measurement result. Since the chord progression is obtained by this, there is an advantage that a natural chord progression whose tonality is controlled can be provided to the user and the chord progression selection work performed by the user can be reduced.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of an automatic chord adding device according to an embodiment of the present invention, FIG. 2 is a flow chart showing the overall operation of the embodiment, FIG. 3 is a diagram showing melody and chord data formats, and FIG. FIG. 6 is a diagram showing an example of melody data stored in the melody data memory, FIG. 5 is a flowchart of reading melody data, FIG. 6 is a diagram showing an example of production data stored in the production rule data memory, and FIG. FIG. 8 is a flowchart for reading production rule data, FIG. 8 is a flowchart for analyzing a melody to determine chords that can be added, and FIG. 9 is a flowchart for calculating a bar start position (Ps) and an end position (pe). , Fig. 10 is a flow chart for Ps calculation, and Fig. 11 is
Flow chart for Pe calculation, FIG. 12 is a flow chart for non-harmonic sound analysis, FIG. 13 is a flow chart for discriminating between harmony and non-harmonic sounds, and FIG. 14 is for calculating a function representing the melody situation before and after the non-harmonic sound. Flow chart, Figure 15
Figure 16, Figure 17, Figure 18, Figure 19, Figure 20, Figure 21, Figure 22
Figure is a detailed flowchart of calculation of each function, Figure 23 is a flowchart of forward inference by production rules,
FIG. 24 is a flowchart for calculating the appearance frequency of chord candidates determined to be addable, FIG. 25 is a flowchart for determining a melody key, FIG. 26 is a flowchart for determining chord progression, and FIG. 27 is a tonality distance. FIG. 28 is a flowchart showing the definition of tonality distance between chords. 1 ... CPU, 2 ... work memory, 3 ... input device, 5
...... Melody data memory, 6 ... Production rule data memory, (R, T) ... Data indicating melody key, dst ₁ ... Measured value of tonality distance between melody tone and chord candidate, dst ₂ ... … Measured tonality distance from previous code, CD
_BAR …… The data of the chord decided for the melody of bar BAR.

Claims (4)

[Claims] 1. An automatic chord adding device for automatically adding chord progression to a given melody, wherein the melody is analyzed to determine chord candidates that can be added to each melody section. Means, distance measuring means for measuring the tonality distance between each melody key and each chord candidate determined to be able to be added to each melody section by the melody analysis means, and each melody analysis means Automatic code addition, characterized by: chord determining means for selecting one chord per melody section from chord candidates determined to be able to be added to the melody section according to the measurement result of the distance measuring means. apparatus. 2. The automatic chord adding device according to claim 1, wherein the melody analyzing means determines that chord candidates which can be added to the melody are entirely distributed.
An automatic chord adding device characterized by further comprising a key discrimination means for discriminating the tone of the melody and giving the information to the distance measuring means. 3. The automatic chord adding device according to claim 1, wherein a first parameter for limiting a tonality distance between the key of the melody and the chord added to each section of the melody, and a parameter before and after the melody. Input means for inputting a second parameter that limits the tonality distance between the preceding chord and the succeeding chord in the section, and the preceding chord candidate for which it is determined by the melody analyzing section that addition is possible for the preceding melody section. Second distance measuring means for measuring the tonality distance between the subsequent chord candidate that can be added to the subsequent melody section, and the chord determining means is the distance measuring means. Of the measurement results of the second distance measuring means, a code giving a measurement result that meets the distance condition specified by the first parameter and the second parameter input from the input means is selected. An automatic code adding device characterized in that 4. An automatic chord adding device for automatically adding chord progression to a given melody, wherein the melody is analyzed to determine chord candidates that can be added to each melody section. Means and the tonality distance between the preceding chord candidate determined to be able to be added to the preceding melody section and the succeeding chord candidate that can be added to the succeeding melody section by the melody analysis means. A chord is selected from the distance measuring means and the chord candidates determined by the melody analyzing means that can be added to each melody section according to the measurement result of the distance measuring means. An automatic code adding device comprising: