JPS61267800A - Performance collation - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for comparing and comparing a model teacher's performance with a student's performance.

Generally, when learning to play a musical instrument, the teacher first performs a model performance, then the student performs the same piece, and the teacher points out mistakes in the student's performance and provides musical guidance. This process is taken.

During this process, there is no need for the teacher to guide in the case of elementary mistakes in performance, such as hitting the wrong pitch or wrong rhythm, and the machine takes over, eliminating the elementary mistakes. At this stage, it is desirable for teachers to provide musical instruction, which is difficult to do with machines. Also, teachers are not always with students, so if a student memorizes a performance incorrectly during self-study, the next time the teacher listens to the student's performance, it will be difficult to correct the mistake. This takes up time and reduces the efficiency of education.

The present invention is a device that allows students to judge for themselves whether or not their own performance is correct, by breaking down the model teacher's performance and the student's performance into musical elements, such as pitch only, rhythm only, Alternatively, it provides an appropriate method for comparing and comparing overall weights.

When comparing two performances, if you want to compare and match single notes, you can compare them in the order in which they were played, but if multiple notes are played at the same time, such as a chord, you should compare them in the order in which they were played. is problematic.

That is, even if the performers intend to play at the same time, there will always be variations in the movements of the performers' fingers and the mechanical actions of the keyboard, resulting in very slight timing deviations.

For example, if the performer is C and E,! :'Even if you try to make the G note sound at the same time and press the keys corresponding to these three notes at the same time, the timing at which the notes actually come out will be slightly different, and the order will be C and E.

G, E, C, G, G.

It may be E or C. This is true for both teacher's and student's performances. Therefore, when comparing chords, it is not possible to compare them in the order in which they were played.

In the present invention, the teacher's performance as a model is referred to as original song data, and the student's performance is referred to as performance data.In the original song data, performance data, or both data, notes pronounced within a certain time range are The following is considered to be a chord:
This is called chord processing. ), other notes are treated as single notes, and when matching the original song data and performance data, chords are matched by following the chronological order for single notes and changing the chronological order for chords. is provided.

According to this method, when beginners practice, they first practice only the pitch, then after learning the pitch they practice the rhythm, then the dynamics, and then the musicality. suitable for proceeding.

That is, when only the pitch is to be compared, the pronunciation timing must be ignored. Therefore, the notes are compared in the order of the played notes. However, as mentioned above, the order in which chords are played may change. In addition, if the original music data is chord-processed and the performance data is compared without chord processing, students can
This is an excellent method that allows you to check the pitches even if you play the notes separately, but not at the same time, and the notes do not form a chord.

The present invention will be explained below based on the drawings.

A circuit configuration for implementing the present invention is shown in FIG.

In the figure, an interface 1 receives performance information from a keyboard or the like, or transmits performance information to a sound source device. There is a MIDI standard as a method for transmitting and receiving performance information, and since it is not essentially related to the present invention, it will not be explained in detail, but the parts related to the explanation of one embodiment of the present invention will be described as follows. Information on key on/off is transmitted as a pair of pitch and key press speed (sound strength), and the pitch is set to center C at key number 60.
The key number is assigned a key number that increases by one for each semitone that goes up, and decreases by one for each semitone that goes down.

The key pressing speed ranges from ■ to 127, with the median value being 64 mf.
(Metuo Forte), and the key pressing speed is O.
It is standardized that # indicates key (key off).

In the embodiment of the present invention, the length of a quarter note is expressed as 120 as a time unit, and this is referred to as time.

The device shown in Figure 1 does not have its own keyboard or sound source, but processes data input through interface 1,
It has functions such as sending data stored in AM4 through an interface and automatically playing other sound sources. Therefore, in the following explanation, the key on/off information or note performance information received by the interface is
This is the method handled by PU2.

The ROM 3 stores a program for controlling the CPU 2. The RAM 4 is a storage device that stores original song data, performance data, verification data, and the like.

Reference numeral 5 denotes a SW section for instructions given to the CPU by a human being, and a display section for displaying processing results.

In explaining the present invention, the musical score shown in FIG. 2 will be used as an example of a teaching material. Each note in the figure is given a note number for convenience of explanation.

First, the teacher performs according to this musical score and creates data for matching the original music.

When the performance information is received by the CPO 2 through the interface 1, the CPU measures the time interval of these received information using an internal counter and makes this correspond to a preset time elapsed value. Convert to time, 3rd
As shown in the diagram in the figure, it is stored in RAM. In the first row of the diagram, note number ■ is key number 72, (center C
C pitch one octave higher than C), key pressing speed 64 (mf
Since this is the first note, the time value is O (that is, the elapsed time value is O).

In the second line, the note number ■ starts from the key-on in the first line.
This indicates that the performance was performed after a time of 240, with a key number of 67 and a key pressing speed of 55.

In the third line, the note number ■ starts from the key-on in the second line.
After time 3, it indicates that the key has been released (key off), and the key number is 72. The key pressing speed is O. (For ease of understanding, key-off is represented by adding a ゛ to the note number, such as °.) Thereafter, the original music data is stored in the same way. When the performance is finished, the CPU stores an end symbol (END) at the end of the data according to the performer's instruction, and then creates the original music matching data chart shown in FIG. 4 based on this data. This process will be explained with reference to the flowchart of FIG. -In the cause,
ST is the step number, m is the line number in Figure 3, l is the length of the note, and t is the time elapsed value of the sounding timing.
This chord time range, which is preset in the judgment unit (203), may vary depending on the performer's technical ability and the difficulty of the piece.
It can be changed, and a range of 10 m sec to 40 m 5 ec in real time is appropriate.

6 is a flowchart for calculating the note length, and FIG. 7 is a flowchart for calculating the elapsed time value of the sound generation timing, which is the time difference from the previous key-on. ) and (202), respectively. In the figure, 1 is the length of the musical note 1m, n is the line number of the chart in FIG. 3, and tm and tn are the time elapsed values of the line numbers m and n, respectively.

The first line of the original music data in FIG. 3 is key-on, and the pitch (key number) is 72.

In order to find out the length of the note, we check the following data and find that the pitch is 67 and the key number is different.When we check the next data, we find that the pitch is 72, the strength is O, the key is off, and the pitch is 67, and the key number is different. Since it matches the pitch of the data in the first line, the time during this time, the time data 240 in the second line, and the time data 3 in the third line add up to 243, which is the length of the note in the first line. Further, the time elapsed value of the sound generation timing is O because it is the first note, and the cumulative time elapsed value is also O. Therefore, as step l, time is 0, pitch is 72, intensity is 64, and length is 2.
43. Enter the cumulative time O in the first line of the chart in Figure 4. Next, when we examine the second line of the original music data chart in Figure 3, we find that this data is also key-on, and the time elapsed value from the first line is 240, and the chord time range is 5 or more. It is determined that step 1 in FIG. 4 is not a chord, and this data is set as step 2. The cumulative elapsed time value is 240. The lengths are calculated in the same manner as in step 1, and the respective values are entered in the second row of the chart in FIG.

The data in the third and fourth rows of the chart in FIG. 3 are of strength O1, that is, key-off, and are therefore not processed.

It can be seen that the fifth line of the chart in FIG. 3 has a strength of 75 and is key-on data. The length is determined in the 1st line.As with the process in the 2nd line, the time elapsed value of the 0 sound timing, which searches for key-offs of the same pitch and calculates the time, is determined by the time difference from the previous key-on, so in the 3rd to 5th lines. Add the time value and get 238
becomes. Since this value is larger than the chord time range 5, it is determined that this note is not the chord of step 2, and is entered as step 3 in the third line of the diagram in FIG. The cumulative time value is 478, which is the sum of the cumulative time value 240 of the previous row and the time value 238.

In the case of the 12th line of the chart in Figure 3, the key is on,
Since the sound generation timing is a time difference of 1 between the previous key-on and the time elapsed value, it is assumed that the chord was sounded at the same time as step 6 in FIG. 4, and is given the same step number. Similarly, the data on the 13th line of the chart in FIG. 3 has the same step number.

Hereinafter, the chart shown in FIG. 4 is created in the same manner.

Next, a method for inputting and processing performance data performed by students will be explained.

In this process, data processing is performed while receiving the student's performance, that is, in real-time processing, the performance data can be converted into performance note information data and compared with the original music comparison data. , how to process the data after inputting it.

The chart in FIG. 8 shows the performance data of the students.

This performance data is first processed as performance note information data according to the flowchart shown in FIG.

In this process, the key-on and key-off of each note are treated as a single step, without being treated as a chord.Other processes are performed using the original song data, including calculation units (201) and (202). The process is similar to that of . The results are shown in the diagram of FIG.

Next, a method for comparing and comparing FIG. 4 and FIG. 10 and determining whether the student's performance is correct or incorrect will be explained.

This process will be explained based on the collation flowchart shown in FIG. In this comparison, only a determination is made as to whether it is correct or incorrect, and the present invention does not describe in detail how to process the result of the determination.

In the figure (401), the line numbers of the original music matching data and performance note information data are set to n and m, respectively, and 1 is set for both as initial values. Also, temporary storage data SF is set to 0. Although it is possible to collate the performance data in the order of the original music data, in this embodiment, the performance data is collated in the order of the performance data. (402) indicates that the verification will stop when the data is completed.

In step (403), the line number of the original song matching data is temporarily stored as On.

In (404), the pitches of the line number m of the performance note information data and the line number n of the original song matching data are compared. If they match, it is judged as positive and the On value is set to 1.
Furthermore, the pronunciation timing, length, and strength are compared and verified, and each element is judged to be correct or incorrect. If the determination is positive at each stage, the next element is further determined, but if it is determined to be an error, the process moves to the first data processing.

By comparing the pitch, then the pronunciation timing, then the length of the note, and then the strength of the note, we can check the basic elements step by step from a music education perspective. In addition, if an error is determined, the subsequent comparison and determination of elements is not performed, thereby improving the processing speed during real-time verification. In addition, length comparison (
It is also possible to omit steps such as 408) and strength comparison (409).

If there is a mismatch in the pitch comparison (404),
(430), and in (420), it is checked whether the note of the original song matching data is a chord. This means that if the step number of the original music matching data is the same as the step number of the next data, it is a chord.

If the next data is a different step number, this data is a single note, so it will be judged as an error (421), and the On value will be set to 1.
is added (422), and the process moves on to the next note.

If the next data has the same step number, it is a chord. SF is temporary storage data for restoring the line number when chord processing is completed, and is 10 while processing a single note.

In (420), if it is determined to be a chord, check the step number of the next note in order, count the number of notes (number of chords) that include that line in the same step, and add n to this value.
The value of is added to obtain SF (431).

1 is added to the line number n of the original music matching data, and the pitches of the performance note information data and the original music matching data with row numbers m and n, respectively, are compared again (434).

If they match, a positive determination (435) is made and other elements are compared and verified. If they do not match, check the step number of the data next to the original music matching data to see if it is a chord. If it is not a chord, it means that the chord does not include a note that matches the pitch of the played note information data, so the determination is erroneous. If the next data is a chord, add 1 to n and check the pitch again.

In this manner, when one piece of performance note information data has been verified, 1 is added to the value of m (411). S
Compare the value of F and the value of m, and if they match, it means that the chord matching is complete, so match the line number n of the original song matching data to the line number m of the performance note information data. , SF
is set to 0 (413).

Thereafter, this process is repeated until the performance note information data is completed.

In the comparison and verification described above, no tolerance range is set for pitch, but tolerance ranges may be set for other pronunciation timing, length, and intensity.

It is desirable to set this tolerance range in several stages depending on the student's ability and the difficulty of the piece.As the student practices, a wide tolerance range is set at first, and then it gradually becomes narrower, i.e., it is judged more strictly. If you do this, you will get a good learning effect.

Based on the original music verification data in Figure 4, here we will provide original music note information with a tolerance range where the key press timing is ±5%, the strength is th15%, and the length is ±10%. The verification data is shown in FIG.

Regarding the data creation in Figure 12, the time, strength, and length in Figure 4 are multiplied by the ratio of their respective tolerance ranges to create verification data with the tolerance ranges in Figure 12.
The data of the allowable range cumulative time value for pronunciation timing comparison is created as follows. That is, for each note data, first, the allowable range time value of the note in FIG. 12 is added to the cumulative time value of the note immediately before the note in FIG. Next, note data in the same step that is considered to be a chord is rewritten to the same allowable range cumulative time value as the first note data in that step. Then, the allowable range cumulative time value in Fig. 12 obtained in this way and the first
Compare with the cumulative time value in Figure 0.

Based on the verification flowchart in FIG. 11, comparing the performance note information data in FIG. 10 with the original music note information verification data in FIG. 12, for line number m=1 in FIG. Compared to row number n=1 in Table 5, the pitch is 72.
It matches 5, and both the strength and length are within the allowable range, so it is determined to be positive. Similarly, m=2 is compared with n=2 and determined to be positive.

m=3 has a different pitch compared to n=3, so
When the next step number n=4 is checked, it is different from the step number n=3, so it is determined to be a single note and a pitch error.

Comparing m=4 with n=4, although the pitches match, the cumulative time is not within the allowable range, so it is determined that the pronunciation timing is incorrect.

For m=5, the pitch, total time, and length are all within the permissible range compared to n=5, but the strength is not within the permissible range, so it is determined that the strength is incorrect.

m=6 has a different pitch compared to n=6, so the next n
When we examine n = 7, we find that the step number is 6, which is the same as the step number for n = 6, so we consider this to be a chord. If we count the number of notes in the chord including n=6, it is 3, so the value of SF is 9 by adding 3 to the value of n, 6.

When we add 1 to n and compare the pitch with n=7, we find that they match. Since the time, length, and strength are also within the allowable range, it is determined to be positive. Return n to On line number 6 in temporary storage (403), and advance m by 1 to 7.

Even in the case of m=7, the pitch is different when compared to the pitch 43 of n=6, so when we look at SF, we find that SF is not O, so when we compare it with the next case of n=7, it is still different. So,
Furthermore, the following n=8 is compared. Since they match here, other elements are checked and it is determined to be correct.

Regarding m=8, when compared again with the pitch of n=6, they match, so other elements are compared. Similarly, since the other elements fall within the allowable range, they are determined to be positive.

Here, in (411) of the flowchart, if 1 is added to m=8, m=9, which matches SF (412
).

Therefore, n is set to the same value as the mouth, the value of SF is set to O, and the next performance data is compared.

Thereafter, determination of correctness or incorrectness is made in the same manner.

In the above process, the performance data is compared with the original song data,
Only the original song data was subjected to chord processing, and the performance data was compared without chord processing. In contrast to this method, it is also possible to perform chord processing on the performance data and then compare the data without performing chord processing on the original song data. In this case, the original music data may be compared with the performance data. Furthermore, it is possible to perform harmonization processing on both the original music data and the performance data and to compare them. The chord time range when performing chord processing on the performance data is preferably larger than the chord time range of the original music data, since the technical ability of the student is lower than that of the teacher.

Furthermore, in the above embodiment, the pitch is determined by MIDI.
Only the key numbers sent by signals are described.

The key number is not only played and output from a keyboard, but also an audio signal such as a guitar or voice may be converted into an electrical signal, the fundamental frequency of its vibration may be extracted, and the key number may be converted into a key number. If the frequency fluctuates due to vibrato, this is averaged and applied to the frequency of the scale. As for pitch information, by deriving and comparing pitch data specified for each note, pitch comparison, which is the most basic stage of music education, can be performed accurately.

In addition, in the above example, the number of notes in the original music data and the performance data are compared on the assumption that they are the same, but in reality, the number of notes in the performance data was mistakenly changed by the student. Errors occur when there are more or less than the number of notes. The processing in this case will be explained based on the flowchart in FIG. In this flowchart, (1) it is assumed that chord processing is being performed together with performance data and original music data, (2) pitch and pronunciation timing are compared and matched, and if an incorrect determination is made, the following procedure is performed. Perform step and verification. In other words, in the middle of the music to be compared, the number of notes in the performance data is greater than the number of notes in the original sound data.
In this embodiment, if there is less than one digit, the location of the error can be detected. Therefore, in this example, if there are two or more missing notes.

(3) In the case of chord matching, even if the number of notes included in the chord of the corresponding performance data is smaller than the number of notes included in the chord of the original sound data, an incorrect judgment will not occur. It is assumed that the

The matching method will be explained below based on the flowchart in FIG. When performing matching, the data to be matched are performance data and original song data, and the lines to be matched are the performance pointer and original song pointer, respectively, and the original song is Match with data. In this embodiment, pitch 9 sound generation timings are compared, but one of each can be omitted. In other words, if pitch comparison is not performed, branch (80
From 2), proceed to branch (820). If you do not want to check the pronunciation timing, branch (820) to (
812).

If you only check the pitch without checking the pronunciation timing, students can ignore the rhythm of the music and check only the pitch of the note, so they can play the pitch accurately. The present invention can provide the most elementary educational stage. Furthermore, if only the pronunciation timing is compared without pitch matching, students can ignore the pitch and only match the musical rhythm, allowing them to create a sense of rhythm that matches the flow of the music. The present invention can provide an educational stage for nurturing. Therefore, the educational effect of the present invention is very large.

In the following description, first, a case will be described in which both pitch comparison and sound generation timing comparison are performed.

In the flowchart of FIG. 13, SSF is a memory that stores the number of step shifts, and the initial setting of SSF is O. Furthermore, the performance T and the original song T are the performance cumulative sound timing, the timing value, and the original song cumulative sound timing value, respectively.

The pitches of the performance data indicated by the performance pointer and the original music data indicated by the original music pointer are compared (803). If they match, the next pronunciation timing is compared (a t O). If the pitches do not match in (803), the step number indicated by the original music pointer is compared with the step number of the next original music pointer in (804). If the step numbers match, it is a chord, so it is compared with the original music data of the blow original music pointer.

If the step numbers do not match, it is not a chord, and the step number of the original song data is advanced. This is called a step shift. Although the limit on the number of step shifts can be set arbitrarily, the limit here is one. If the number of step shifts is less than the limit, perform a step shift and move the original pointer to the beginning of the next step.808
), check again. In this case, omitting the sound is incorrect (901
), and if the step shift limit is exceeded, the sound is determined to be incorrect (902). Calculation of the number of step shifts is performed in (805), and the process is returned to O in (809).

If the pitches match in the pitch comparison (803), the next generation timing comparison, 0 generation timing, is performed within a certain tolerance range. If it does not fall within this tolerance range and the cumulative sound timing of the original song data is earlier (value ゛ is smaller) than the cumulative sound timing of the performance data, the omission is incorrect (904) and the step of the original song data is earlier than the cumulative sound timing of the performance data. If the cumulative pronunciation timing of the original song data is later (larger value) than the cumulative pronunciation timing of the performance data, the timing of the performance was too early or there were too many notes. Therefore, it is considered a mistake.

When the comparison is completed, the pointer of the performance data is advanced, and the original pointer is next aligned with the note number of the original music data to be compared.

The above matching method will be explained based on the musical score example shown in FIG. In the figure, (t o i) is the original song, and (102) is the student's performance. The diagram in FIG. 15 shows the above (lot), (
102) are replaced with original music note information verification data and performance note information data, respectively. In this embodiment, the allowable range of sound generation timing is set to ±10%. The method of setting each data is the same as that described in FIGS. 10 and 12.

Performance pointer 1.2 and original song pointer l.

2 have the same pitch 9 sound timing. Comparing the performance pointer 3 and the original music pointer 3, the pitches are different. If we check the step number of pointer 4 next to the original song data, it is different from step 3, so it is not an original chord.

Therefore, the step shift is set to 1, the step of the original music data is advanced by 1 to become step 4, and the pointer is placed on the original music pointer 4 where the first note number of step 4 is located for comparison. This still does not match, so when we check the step number of the next pointer in the original music data, we find that it is the same as step number 4 of original music pointer 4, so it is a chord, so we move the original music pointer without performing a step shift. Go ahead and compare.

Although this still does not match, the step number of the next original music pointer 6 is also the same step number 4, so when the original music pointer is advanced one more time to the original music pointer 6 and compared, the pitches match.

If you check the pronunciation timing, it will also match, so point 3 to the original song data. Note number ■ is determined to have been skipped during performance.

Error display (901) is for displaying missing notes.

Next, advance the performance pointer to 4, and check the step number of performance pointer 4 at (813), and find that the previous step number - 3
Since it is the same as and has not changed, it is a played chord, and in this case, the original pointer is returned to pointer 4, which is the first note number in the same step 4.

Then, when comparing the pitches, they match, and the sound timing also matches, so the performance pointer is advanced to 5. At this time, the step number of the performance data has changed to 4, so it is not a performance chord, but the original pointer of the original song data advances the step in the original song data by 1 and points to the original pointer 7 where the first note number of step 5 is. is set to

In this embodiment, the note number ■ in the original music data is either absent or sung in the performance data, but this is not displayed incorrectly. This means that when a piano piece contains many notes in the same chord, it may be necessary to omit notes that are difficult to play and are not important (such as octave notes or fifths). In this case, if all errors are displayed when compared with the original song data, it is not desirable because it will weaken the impression of the incorrect display of other notes. Also, if the original song data is a melody that contains double notes, the performance data will not be correct. The highest note, the melody
This produces an excellent educational effect, as no erroneous indications will appear even if only one note is played.

Next, when the performance pointer 5 and the original music pointer 7 are compared, the pitch 1 generation timings are the same. Performance pointer 6
and the original music pointer 8 also match.

Comparing the performance pointer 7 and the original music pointer 9, the pitches are different. When we check the step number of the pointer next to the original pointer, it is not step 7, so it is not an original chord.Even if we perform a step shift, set the original pointer to 10, and compare the pitches, the pitches are still different. Furthermore, when the step number of the pointer next to the original music pointer is checked, the step number is not 8, so a step shift is attempted. However, since the step shift limit is 1, no step shift is performed and a performance error is determined (902). In this case, the performance pointer is then advanced by 1 to 8, and the original music pointer is set back by 1 from the step number 8 of the original music data, and is set to the original music pointer 9 where the first note number of step 7 is located.

Comparing the performance pointer 8 and the original pointer 9, the pitch is 1.
The pronunciation timing also matches.

The performance pointer 9 and the original music pointer 10 also match.

If you compare the performance pointer 10 and the original song pointer 11, the pitches will match, but when you compare the pronunciation timings, they will be different. Moreover, the cumulative pronunciation timing value of the performance is less than the cumulative pronunciation timing value of the original song. A performance error occurs, and in this case, only the performance pointer is advanced by 1 to 11.

Comparing the performance pointer 11 and the original pointer 11, it is found that the timing of pitch 9 is also the same. □In this way, the matching ends. If the matching is 0 or more, an error is displayed (902
) and (903) display the remaining sound.

Based on the same flowchart in FIG. 13, when only pitch is to be verified, the process proceeds from branch (820) to (812). Therefore, the rhythm during the performance shown in FIG. 15 is ignored, and missing notes (note number ■ of the original song), extra notes (note number ■ of the performance), and extra notes (note number 0 of the performance) are displayed incorrectly. If only one sound generation timing is to be verified, the process proceeds from branch (802) to (820). Therefore, pitch errors are ignored, and rhythm omissions (original note number ■), rhythm excess (playing note number ■), and rhythm excess (note number ■ of the original song) are ignored.
The note number (IF) of the performance is displayed incorrectly.

As explained above, according to the matching method according to the present invention, a group of notes pronounced within a certain short period of time, such as a chord, is regarded as a chord, and when matching chords, the chronological order is changed. By doing this, it is possible to match appropriate chords. In addition, during matching, musical elements are broken down into pitches, sound timings, note lengths, and sound strengths, making it possible to match only pitches, sound timings only, and even sound timings, note lengths, and sound strengths. , by setting acceptable ranges for sound intensity in stages,
It has an excellent educational effect as it makes it possible to compare elements according to the stage of music education.

[Brief explanation of the drawing]

FIG. 1 is an example of a circuit configuration used to carry out the method of the present invention, FIG. 2 is an example of a musical score used to explain the present invention in detail, and FIG. 3 is a diagram representing original music data. Figure 4 is a diagram showing the data for matching the original music, Figure 5 is a flowchart used for determining whether the original music is a chord or a single note, or for determining the C note, and Figure 6 is the note length. Figure 7 is a flowchart for calculating the pronunciation timing, Figure 8 is a diagram showing performance data, Figure 9 is a flowchart for processing without chord processing, Figure 10 is performance note information data. Figure 11 is a flowchart used to compare note information, Figure 12 is a diagram showing data for comparing original music note information, and Figure 13 is a flowchart used when the number of notes does not match between the original music and the performance. FIG. 14 is an example of the musical score used for the explanation of FIG. 13, and FIG. 15 is a chart in which the original music and the student's performance in FIG. 14 are replaced with data. Patent Applicant: Roland Co., Ltd. Representative Iku Kadashi Tabu Gaku Engraving of the drawing in Figure 2 of the score (no changes to the content) Figure 4 Diagram showing the data for checking the original piece Figure 5 Procedures Amendment May 1985 27th, 1985 Patent Application No. 11468
1. Name of the invention
2. Performance matching method. 3. Relationship with the case of the person making the amendment: Patent applicant Postal code: 559 Address: 3-7-13-4 Shinkokujima, Suminoe-ku, Osaka City;
Date drawing of amendment order. 6. Contents of amendment
6 Fig. 3, Fig. 4, Fig. 8, Fig. 1O, Fig. 12. Figure 15 has been corrected into a proper drawing using dark ink as shown in the attached sheet. Procedural amendment June 18, 1986 Mr. Michibu Uga, Commissioner of the Patent Office Relationship with the case: Patent applicant Postal code 559 Address 3-7-13 Shinkitajima, Suminoe-ku, Osaka City Amendment order Date: May 27, 1986. Contents of the amendment in the name of the invention column of the specification to be amended. "Performance matching method" in the third line of page 1 of the specification is amended to "performance matching method."

Claims (1)

[Claims] 1) A performance matching method in which original music data and performance data are input, individual note information contained in the data is made to correspond with each other, and the correctness or incorrectness of each note in the performance data is detected, For at least one of the original song data and the performance data, a process is performed to determine that a group of notes included in a preset time range are chords based on the pronunciation timing information of each note, and the notes included in the time range are determined to be chords. Notes that do not match are processed to be determined as single notes, and the above chord and/or single note are made to correspond in chronological order.Furthermore, when each note included in the chord is made to correspond, the chronological order is determined to be different. A performance verification method in which each note is compared and verified to determine whether the performance is correct or incorrect. 2) The performance matching method according to claim 1, wherein the preset time range is variable. 3) The performance matching method according to claim 1, wherein the preset time range is set to be wider when determining performance data than when determining original music data. 4) In a performance matching method that inputs original music data and performance data, matches individual note information contained in the data, and detects whether each note in the performance data is correct or incorrect, one of the original music data or From the performance data, pitch matching data derived as data specified for each note based on the pitch information of each note, and data for pitch matching derived as data specified for each note based on the pitch information of each note, and data for each note based on note information other than the above pitch information of each note. The note matching data is derived as data having a preset tolerance range for each note, and the note information matching data is created based on the note information of each note from the other performance data or original song data. The musical note information data consisting of the pitch data derived from the pitch data and the note data other than the pitch is compared and verified with the note information verification data to determine whether each note in the performance data is correct or incorrect. Method. 5) The performance matching method according to claim 4, wherein the note matching data includes data corresponding to at least one of note pronunciation timing information, note length information, and note strength information. . 6) The performance matching method according to claim 4, wherein the note matching data allows setting at least two or more different allowable ranges for each note information of each note. 7) A patent claim that makes it possible to select and match either or both of pitch matching data and data corresponding to note pronunciation timing information from among musical note information matching data. The performance matching method described in Section 4. 8) When at least the pitch matching data, the data corresponding to the note pronunciation timing information, and the data corresponding to the note length information among the note information matching data are compared and matched with the note information data. , the pitch data of the note is first compared, and only when a correct performance judgment is made, the data corresponding to the pronunciation timing information of the note and the data corresponding to the length information of the note are compared and verified. The performance matching method described in Section 4. 9) Among the note information data, the pitch data is first compared and verified, and only when the performance is determined to be correct, the data corresponding to the pronunciation timing information of the note is compared and verified, and the performance is determined to be correct. Only when a determination is made, the data corresponding to the length information of the note is compared and verified, and only when the performance is correctly determined, the data corresponding to the strength information of the note is compared and verified. Performance matching method described in Scope 4. 10) Compare and match the note information matching data and note information data in chronological order, and if an incorrect performance judgment is made, the processing order is changed within a predetermined range in the chronological order. Claim 4, in which the musical note information data of the performance data for which the incorrect determination has been made is again compared and verified with the musical note information collation data of the corresponding original music data.
Performance matching method described in section. 11) Claim 1 which does not make an erroneous determination even when the number of notes constituting a chord in the performance data is smaller than the number of notes constituting a chord in the corresponding original song data.
Performance matching method described in section.