JP2021156975A - Tempo detector, method, and program - Google Patents

Abstract

To more accurately detect a tempo on technique for detecting the tempo from waveform data of music.SOLUTION: In steps from a step S101 to S104, music waveform intensity data indicating distribution of waveform intensity is acquired from waveform data of music. In a step S105, autocorrelation on the basis of the music waveform intensity data is detected, and a basic autocorrelation value indicating intensity for each period is calculated. In the step S105, an autocorrelation value for calculation, which is obtained by changing a period of the basic autocorrelation value by a magnification ratio determined on the basis of a designated rhythm, is calculated. In a step S106, a portion where an autocorrelation value for detection, which is obtained by weighting and adding the autocorrelation value for calculation to the basic autocorrelation value, becomes the maximum is specified and a period corresponding to the specified portion is detected as a tempo of the music.SELECTED DRAWING: Figure 1

Description

The present invention relates to a technique for detecting a tempo from waveform data of a musical piece.

When detecting the tempo of a musical piece from the musical piece waveform data, a technique for determining the tempo of the musical piece is known based on the periodicity calculated by the autocorrelation calculation for the input data related to the musical piece waveform data (for example). , Patent Document 1).

JP-A-2002-116754

However, in the above-mentioned conventional technique, it is difficult to accurately detect the tempo when the intensity change peculiar to the music is included.

Therefore, an object of the present invention is to detect the tempo more accurately.

In one example of the embodiment, music waveform intensity data indicating the distribution of waveform intensity is acquired from music waveform data, autocorrelation based on music waveform intensity data is detected, and a basic autocorrelation value indicating intensity for each cycle is calculated. The calculation autocorrelation value is calculated by specifying the beat corresponding to the music and changing the period of the basic autocorrelation value at a magnification determined based on the specified beat, and the calculation autocorrelation value with respect to the basic autocorrelation value. A control unit is provided that identifies the portion where the autocorrelation value for detection is maximized by weighting and adding the above, and detects the period corresponding to the identified portion as the tempo of the music.

According to the present invention, the tempo can be detected more accurately.

It is a processing block diagram of a tempo detection method. It is a graph which shows the example of the beat waveform data, the autocorrelation value, and the weighted autocorrelation value. It is a processing block diagram of the timing synchronization method. It is a graph which shows the example of the timing level in the music example of 3 beats. It is a graph which shows the example of the timing level in the music example of 4 beats. It is a block diagram which shows the hardware configuration example of the computer which executes the process of tempo detection and timing synchronization. It is a main flowchart which shows an example of tempo detection / timing synchronization processing. It is a flowchart which shows the detailed example of a tempo detection process. It is a flowchart which shows the detailed example of a tempo determination process. It is a flowchart which shows the detailed example of the timing synchronization processing and the time signature detection processing. It is a flowchart which shows the detailed example of the synchronous offset time detection processing.

Hereinafter, embodiments of the tempo detection method and the timing synchronization method for carrying out the present invention will be described in detail with reference to the drawings. FIG. 1 is a processing block diagram of the tempo detection method in the present embodiment. The tempo detection method is
Step S101: Voice input processing Step S102: Envelope detection processing Step S103: Beat waveform detection processing Step S104: Beat intensity detection processing Step S105: Autocorrelation detection processing Step S106: Tempo determination processing.

First, in the audio input process of step S101, the audio input from the microphone input device or the like is A / D (analog / digital) converted at a constant sampling cycle (for example, 44.1 KHz (kilohertz)), and the music waveform data is converted. get.

Next, in the envelope detection process of step S102, the music waveform data acquired by the voice input process of step S101 is divided into blocks for each fixed number of samples (for example, 128 samples), and the absolute music waveform data is absolute for each block. Calculate the music waveform intensity, which is the average of the values. Further, the envelope detection process executes a weak (high cutoff frequency) low-pass filter process on the calculated music waveform intensity, and acquires an envelope waveform as a calculation result of the low-pass filter process.

Next, in the beat waveform detection process of step S103, a strong (low cutoff frequency) low-pass filter process is executed on the music waveform intensity calculated by the envelope detection process of step S102, and the samples having the same sampling time are used. Then, by subtracting the calculation result of the strong low-pass filter processing from the envelope waveform, the beat waveform is acquired as the subtraction result.

FIG. 2A is a graph showing an example of a beat waveform. The part of the bar graph 201 is the envelope waveform. The part of the line graph 202 shows the result of executing the low-pass filter processing strong against the music waveform intensity, and the line graph 203 is obtained by subtracting the result 202 of the strong low-pass filter processing from the envelope waveform 201 between the corresponding samples. It is a beat waveform (double magnification). It is understood that the beat waveform is the result of removing the standing wave portion having a low correlation with the tempo from the envelope waveform.

In the beat intensity detection process of step S104, a low-pass filter process is executed on the absolute value of the beat waveform calculated in step S103, and the beat intensity is calculated as the processing result.

Here, the tempo of the music is expressed by the unit of the number of beats per minute: bpm (beats per minute), and is expressed as "the tempo is 60 bpm." This indicates the tempo (speed) of a musical piece in which 60 beats are counted per minute. Then, in FIG. 1, the autocorrelation detection process in step S105 calculates the autocorrelation value (basic autocorrelation value) of the beat waveform at the time length of one bar of the music calculated from the bpm value of the tempo. For example, if the music is in 4 beats and the tempo value is BPM, the time length TL (seconds) for one measure is
TL = 60 × 4 / BPM
Will be. Then, assuming that the number of samples in one block when calculating the waveform intensity of the music is SPB and the sampling period (number of samples / second) is SPR, the number of blocks BL for one bar in the beat waveform is
BL = SPR x TL / SPB
Is calculated as. Then, in step S105, the autocorrelation value of the beat waveform at the BL interval is calculated.

FIG. 2B is a graph showing an example of the autocorrelation value calculated in step S105, and is an example of an analysis result of a dance music song having a tempo value of 130 bpm. The horizontal axis is the bpm value, and the detection range is from 60 bpm to 180 bpm. In addition to the position 204 of the original tempo value of 130 bpm (4 beat time length) of the music, a strong correlation appears at positions such as 104 bpm (5 beat time length) and 173 bpm (3 beat time length), and the position 205 of 173 bpm. Is the maximum value of the autocorrelation value in this graph, and the tempo cannot be detected accurately as it is.

Therefore, in the tempo determination process of step S106 of FIG. 1, the autocorrelation values of 4/5 times bpm and 4/3 times bpm (calculation) are added to the autocorrelation values of each bpm position calculated by the autocorrelation detection process of step S105. The autocorrelation value for detection) is weighted and added, and the bpm value at which the autocorrelation value (autocorrelation value for detection) obtained as a result of this weighting addition is maximized is determined as the tempo.

Now, assuming that the autocorrelation value at a certain bpm position is ACR [bpm], the weighted autocorrelation value WACR [bpm] is
Calculation example 1: WACR [bpm] = ACR [bpm]
+ ((ACR [bpm x 4/5]]
+ ACR [bpm x 4/3]) / 2
-ACR [bpm]) x WCF
Calculation example 2: WACR [bpm] = ACR [bpm]
+ ((ACR [bpm x 4/5]]
+ ACR [bpm x 4/3]) / 2 x WCF
Will be. If bpm x 4/5 is less than the lower limit of the bpm detection range,
WACR [bpm] = ACR [bpm]
+ (ACR [bpm x 4/3]
-ACR [bpm]) x WCF
If bpm x 4/3 exceeds the upper limit of the detection range of bpm,
WACR [bpm] = ACR [bpm]
+ (ACR [bpm x 4/5]
-ACR [bpm]) x WCF
And.
Here, WCF is a weighting coefficient of 0.0 or more and 0.5 or less, which is determined based on the beat intensity calculated in step S104. For example, if the beat intensity (hereinafter, this value is referred to as "BS") takes a value from 0 to 4095,
If BS is 1024 or less, WCF = 0.0
If BS is 3096 or more, WCF = 0.5
If BS is other than the above, WCF = (BS-1024) / 4096
And so on.

FIG. 2C is a graph showing an example of autocorrelation values weighted and added in step S106, and is an example of analysis results of the same music as in FIG. 2B. The self-correlation value at the 4/5 times bpm position (5 beat time length) and the 4/3 times bpm position (3 beat time length) is suppressed, and the original tempo value of the song is 130 bpm (4 beat time length) at position 206. It is understood that the tempo detection was performed accurately because the position 207 where the autocorrelation value had the maximum value coincided with the position 207.

In the tempo determination process described above, the autocorrelation values of the 4/5 times bpm position and the 4/3 times bpm position were weighted and added on the premise of 4 beats, but it was determined to be 3 beats by the beat detection process in the timing synchronization method described later. If so, the autocorrelation values of the 3/4 times bpm position and the 3/4 times bpm position are weighted and added.

FIG. 3 is a processing block diagram of the timing synchronization method in the present embodiment. The timing synchronization method is
Step S301: Beat detection process (calculation of 1 beat time timing level)
Step S302: Synchronous offset time detection process (1 bar timing level calculation)
Consists of.

First, in the time signature detection process of step S301, the timing level is detected by repeating the time signature of 1/3 (3 beats) and 1/4 (4 beats) of one bar obtained by the tempo detection process of FIG. do.

In step S301, the timing level is a value obtained by subtracting the previous value from the current value of the beat waveform (for example, 203 in FIG. 2A) detected by the beat waveform detection process in step S103 of FIG. If it is negative, the value is 0, and if it is positive, it is a value obtained by applying a low-pass filter to the value for each beat time resolution.

In step S301, for example, assuming that the time resolution of one beat is 96, the time length of one bar is decomposed into the timing of 96 × 3 = 288 for 1/3 hour length (3 beats), and 1/4 hour length (4 beats). Then, the time length of one bar is decomposed into the timing of 96 × 4 = 384, and each timing level value for each time resolution (96) of one beat is calculated.

Then, in step S301, the magnitude of the difference between the maximum value and the minimum value of the obtained timing level is compared, and the time signature is determined. That is, if the maximum value-minimum value of the timing level value of 1/3 hour length is larger than the maximum value-minimum value of the timing level value of 1/4 hour length, the music is determined to be in triple time. NS. On the contrary, if the maximum value-minimum value of the timing level value of 1/4 hour length is larger than the maximum value-minimum value of the timing level value of 1/3 hour length, it is determined that the music is in 4 beats. Will be done.

In the time signature detection process of step S301, the time signature of the music may be determined by the deviation of the timing level. Further, 5 beats, 7 beats and the like can be detected by the same method as described above.

FIG. 4 is a graph showing an example of the timing level detected by the time signature detection process in step S301 of FIG. 3 with respect to an example of a musical piece having three beats. 4 (a) is 1 bar length, FIG. 4 (b) is 1/2 bar length, FIG. 4 (c) is 1/3 bar length, and FIG. 4 (d) is 1/4 bar length. The maximum value-minimum value of the timing level value in the 1/3 hour length of one bar shown in FIG. 4 (c) is the maximum value-minimum value of the timing level value in the 1/4 hour length shown in FIG. 4 (d). Since it is larger than the value, it is determined that the music has three beats, and the beat of the music is detected correctly.

FIG. 5 is a graph showing an example of the timing level detected by the time signature detection process in step S301 of FIG. 3 with respect to an example of a four-beat music piece. 5 (a) is 1 bar length, FIG. 5 (b) is 1/2 bar length, FIG. 5 (c) is 1/3 bar length, and FIG. 5 (d) is 1/4 bar length. The maximum value-minimum value of the timing level value in the 1/4 hour length of one bar shown in FIG. 5 (d) is the maximum value-minimum value of the timing level value in the 1/3 hour length shown in FIG. 5 (c). Since it is larger than the value, it is determined that the music has four beats, and the beat of the music is detected correctly.

Returning to the description of FIG. 3, the synchronous offset time detection process of step S302 is performed for each bar time length (see the description of step S104 of FIG. 1) calculated from the tempo bpm value detected by the tempo detection process of FIG. Detects the timing level.

In step S302, for example, assuming that the resolution of one beat is 96, when the result of the time signature detection process of step S301 in FIG. 3 is three beats, the time length of one bar is decomposed into the timing of 96 × 3 = 288, which is the timing. When the result is four beats, the time length of one bar is decomposed into timings of 96 × 4 = 384, and the timing level value for each timing is calculated.

Then, in step S302, the timing at which the timing level value obtained as described above becomes the maximum value is detected as the synchronization offset time. Here, the synchronization offset time is the start time of the beginning of one bar, which is the original of the music, in the timing level detection processed by repeating the time length of one bar.

Further, when the time signature detection process of step S301 in FIG. 3 determines that the music has four beats, the synchronization offset time detection process of step S302 detects the timing level at a length of 1/2 hour of one bar. The timing T1 at which the level value becomes the maximum value and the timing level value at the timing T2 at which 1/2 hour of one bar is added from T1 are compared, and the larger timing level value is synchronized offset. It may be time. With such an algorithm, it is possible to reduce the possibility of erroneously detecting the synchronization offset time from an accent that constantly enters the back of the 4th beat.

In the synchronization offset time detection process of step S302 described above, the synchronization unit is set to one bar time length, but when the synchronization unit is set to one beat time length, the timing is set to one beat time length in the beat detection process of step S301. The timing at which the level value becomes maximum may be the synchronization offset time.

Since the beat / bar timing synchronization method of FIG. 3 executes the process based on the time length of one bar detected by the tempo detection process of FIG. 1, if the detected tempo changes, the process will be redone. Therefore, there is a problem that it takes time to complete the synchronization.

Therefore, the beat waveform obtained by the beat waveform detection process in step S103 of the tempo detection process of FIG. 3 is buffered for the bar time of n (n is an integer of 2 or more), which is the lowest tempo value that can be detected. By re-executing the processing for n measures of the new tempo at the timing when the tempo changes, using the buffered data, the synchronization can be resumed promptly.

When it is determined in the beat detection process of step S301 of FIG. 3 that the music has 3 beats, the detection unit of the autocorrelation detection process of the tempo detection method of FIG. 1 in the present embodiment is tempo bpm assuming 4 beats. Since it is a value, the detected bpm value shall be multiplied by 3/4 and reported.

FIG. 6 is a diagram showing an example of a computer hardware configuration capable of realizing the processing block of the tempo detection method of FIG. 1 and the processing block of the timing synchronization method of FIG. This computer includes a smartphone, a tablet terminal, a digital camera, etc. in addition to a normal personal computer. The computer shown in FIG. 6 includes a CPU (Central Processing Unit) 601, a memory 602, an input device 603, an output device 604, an auxiliary information storage device 605, a medium drive device 606 into which a portable recording medium 609 is inserted, and a network connection. It has a device 607. These components are connected to each other by bus 608. The configuration shown in FIG. 6 is an example of a computer capable of realizing the processing block of the tempo detection method of FIG. 1 and the processing block of the timing synchronization method of FIG. 3, and such a computer is not limited to this configuration.

The memory 602 is a semiconductor memory such as a Read Only Memory (ROM), a Random Access Memory (RAM), or a flash memory, and stores a program and data used for processing.

The CPU (processor) 601 (control unit) is used, for example, in the processing block of the tempo detection method of FIG. 1 and the processing block of the timing synchronization method of FIG. 3, using the memory 602, for example, the flowcharts of FIGS. 8 to 11. By executing the program corresponding to the processing of FIG. 1, it operates as each processing block of FIGS. 1 and 2.

The input device 603 is, for example, a keyboard, a pointing device, or the like, and is used for inputting an instruction or information from an operator or a user. The output device 604 is, for example, a display device, a printer, a speaker, or the like, and is used for making an inquiry to an operator or a user or outputting a processing result.

The auxiliary information storage device 605 is, for example, a hard disk storage device, a magnetic disk storage device, an optical disk device, a magneto-optical disk device, a tape device, or a semiconductor storage device. The processing block of the tempo detection method of FIG. 1 and the processing block of the timing synchronization method of FIG. 3 are used in the auxiliary information storage device 605 to process the flowcharts of FIGS. The program and data to be executed can be stored and loaded into the memory 602 for use.

The medium driving device 606 drives the portable recording medium 609 and accesses the recorded contents. The portable recording medium 609 is a memory device, a flexible disk, an optical disk, a magneto-optical disk, or the like. The portable recording medium 609 may be a Compact Disk Read Only Memory (CD-ROM), a Digital Versaille Disk (DVD), a Universal Social Bus (USB) memory, or the like. The operator or the user can store the above-mentioned program and data in the portable recording medium 609 and load it into the memory 602 for use.

As described above, the computer-readable recording medium for storing the above-mentioned programs and data is a physical (non-temporary) recording medium such as a memory 602, an auxiliary information storage device 605, or a portable recording medium 609. Is.

The network connection device 607 is a communication interface that is connected to a communication network such as Local Area Network (LAN) and performs data conversion associated with communication. The computer of FIG. 6 can receive the above-mentioned programs or data from an external device via the network connection device 607, load them into the memory 602, and use them.

It is not necessary for each of the processing blocks of FIGS. 1 and 3 to include all the components of FIG. 6, and some components may be omitted depending on the application or conditions. For example, the input device 603 may be omitted if it is not necessary to input instructions or information from the operator or user. When the portable recording medium 609 or the communication network is not used, the medium driving device 606 or the network connecting device 607 may be omitted.

FIG. 7 is a main flowchart showing an example of tempo detection / timing synchronization processing executed by a computer having the configuration example of FIG. In FIG. 6, this process is a process in which the CPU 601 loads the tempo detection / timing synchronization processing program stored in the auxiliary information storage device 605 or the like into the memory 602 and executes it. Hereinafter, description will be made with reference to FIG. 6 as needed.

In FIG. 7, the CPU 601 first initializes the tempo variable value to 0 (undecided), the time signature variable value to the value 4 indicating four beats, and the synchronization offset time variable value to 0 (step S701). These variable values are stored in the memory 602.

After that, the CPU 601 repeatedly executes a series of processes from step S702 to step S705.

In the above series of processes, the CPU 601 first executes a tempo detection process, calculates and sets an undetermined value 0 or a value between 60 bpm and 180 bpm in the tempo variable stored in the memory 602 (step S702). ). A detailed example of this process will be described later with reference to FIG.

In the above series of processes, the CPU 601 then determines whether or not the user has instructed the end of the tempo detection / timing synchronization process by a switch (not shown) or the like that constitutes the input device 603 (step S703). ).

If the determination in step S703 is YES, the CPU 601 ends the tempo detection / timing synchronization process shown in the main flowchart of FIG. 7, and returns to the process of the control program, which is not particularly shown.

If the determination in step S703 is NO, the CPU 601 determines whether or not the tempo variable value on the memory 602 detected by the tempo detection process in step S702 is not 0 (step S704).

When the tempo variable value is an undecided value 0 (the determination in step S704 is NO), the CPU 601 returns the control to step S702 and continues repeating the series of processes described above.

When the tempo variable value is other than the undecided value 0 (the determination in step S704 is YES), the CPU 601 executes the timing synchronization process to add 3 beats, 4 beats, and the beat variable stored in the memory 602. Alternatively, the values 3, 4, or 5 indicating any of the five beats are acquired, and any of the values 0 to n (n is a natural number) is calculated in the synchronization offset time variable stored in the memory 602. Set.

After that, the CPU 601 returns the control to step S702 and continues the repetition of the series of processes described above.

FIG. 8 is a flowchart showing a detailed example of the tempo detection process in step S702 of FIG. Here, the CPU 601 executes the same tempo detection process as described in each process block of FIG.

That is, first, the CPU 601 executes the same voice input process as in step S101 of FIG. 1 (step S801). In the process of step S801, the CPU 601 inputs the sound input by the user from the microphone input device or the like that constitutes a part of the input device 603, and A / D (analog / analog /) (not particularly shown) that also constitutes a part of the input device 603. The digital) converter performs A / D conversion at a constant sampling period (for example, 44.1 KHz (kilohertz)), acquires music waveform data, and stores it in the memory 602.

Next, the CPU 601 executes an envelope detection process similar to step S102 in FIG. 1 (step S802). In the process of step S802, the CPU 601 divides the music waveform data acquired by the voice input process of step S801 and stored in the memory 602 into blocks for each fixed number of samples (for example, 128 samples), and music for each block. The music waveform intensity data, which is the average of the absolute values of the waveform data, is sequentially calculated and stored in the memory 602. Further, the CPU 601 sequentially reads out the music waveform intensity data calculated and stored in the memory 602, executes a weak (high cutoff frequency) low-pass filter processing, and results in an envelope as a calculation result of the low-pass filter processing. Waveform data is sequentially acquired and stored in the memory 602.

Next, the CPU 601 executes the same beat waveform detection process as in step S103 (step S803). In the process of step S803, the CPU 601 sequentially reads the music waveform intensity data calculated in the envelope detection process of step S802 and stored in the memory 602, and executes a strong (low cutoff frequency) low-pass filter process. By subtracting the sample of the calculation result data of the above strong low-pass filter processing from the sample of the envelope waveform data sequentially read from the memory 602 among the samples having the same sampling time, the beat waveform data is sequentially acquired as the subtraction result. , Stored in memory 602.

Subsequently, the CPU 601 executes the beat intensity detection process similar to the step S104 of FIG. 1 (step S804). In the process of step S804, the CPU 601 sequentially reads the beat waveform data calculated in step S803 and stored in the memory 602, calculates the absolute value of each sample, executes low-pass filter processing on the absolute value, and performs the process. As a result, the beat intensity data is calculated and stored in the memory 602.

Subsequently, the CPU 601 executes an autocorrelation detection process similar to step S105 in FIG. 1 (step S805). In the process of step S805, the CPU 601 calculates the autocorrelation value of the beat waveform at the time length of one bar of the music calculated from each bpm value in the assumed tempo range.

Then, the CPU 601 executes the tempo determination process similar to the step S106 of FIG. 1 (step S806). After the process of step S806, the CPU 601 ends the tempo detection process of step S702 of FIG. 7 shown in the flowchart of FIG.

FIG. 9 is a flowchart showing a detailed example of the tempo determination process in step S906 of FIG.

In FIG. 9, the CPU 601 first determines whether or not the value of the time signature variable set in the memory 602 is 3 by the user designating the time signature by a switch forming a part of the input device 603. (Step S901).

If the determination in step S901 is YES, the CPU 601 sets the autocorrelation value of each bpm position calculated in the autocorrelation detection process of step S805 of FIG. 8 and stored in the memory 602 to 3/2 times bpm and 3 /. The autocorrelation value of 4 times bpm is weighted and added (step S902).

If the determination in step S901 is NO, the CPU 601 determines whether or not the value of the time signature variable set in the memory 602 is 5 by the user designating the time signature by a switch forming a part of the input device 603. (Step S903).

If the determination in step S903 is YES, the CPU 601 increases the autocorrelation value of each bpm position calculated by the autocorrelation detection process in step S805 of FIG. 8 and stored in the memory 602 by 5.4 times bpm and 5 /. The autocorrelation value of 6 times bpm is weighted and added (step S904).

If the determination in step S903 is NO, the value of the time signature variable set in the memory 602 is set to 4 by the user designating the time signature by a switch forming a part of the input device 603. In this case, the CPU 601 has 4/3 times bpm and 4/5 times bpm self in the autocorrelation value of each bpm position calculated by the autocorrelation detection process in step S805 of FIG. 8 and stored in the memory 602. The correlation values are weighted and added (step S905).

After the processing of steps S902, S904, or S905, the CPU 601 detects the bpm value that maximizes the autocorrelation value obtained as a result of the weighting addition in the processing of steps S902, S904, or S905 (step S906).

Then, the CPU 601 updates the bpm value detected in step S906 as a new tempo value by replacing the value of the tempo variable stored in the memory 602 with the above bpm value (step S907).

After that, the CPU 601 ends the tempo determination process of step S806 of FIG. 8 shown in the flowchart of FIG.

FIG. 10A is a flowchart showing the timing synchronization process of step S705 of FIG. 7. In this timing synchronization process, the CPU 601 first executes the same time signature detection process as in step S301 of FIG. 3 (step S1001), and then executes the same synchronization offset time detection process as in step S302 of FIG. 3 (step S1002). ), After that, the process of step S705 of FIG. 7 shown in the flowchart of FIG. 10 is completed.

FIG. 10B is a flowchart showing a detailed example of the time signature detection process of FIG. 10A. In FIG. 10B, the CPU 601 first has a time length of 1/3 (3 beats) of 1 bar obtained by the tempo detection process of step S702 of FIG. 7, as described in step S301 of FIG. The timing level is detected and stored in the memory 602 by repeating the above (step S1011).

In FIG. 10B, the CPU 601 then has a time length of 1/4 (4 beats) of 1 bar obtained by the tempo detection process in step S702 of FIG. 7 as described in step S301 of FIG. The timing level is detected and stored in the memory 602 (step S1012).

In FIG. 10B, the CPU 601 then repeats the timing in the same manner as above with a time length of 1/5 (5 beats) of 1 bar obtained by the tempo detection process of step S702 in FIG. The level is detected and stored in the memory 602 (step S1013).

After that, the CPU 601 detected the timing level value having the largest deviation (or the maximum value-minimum value described in step S301) of each timing level value detected in steps S1011, S1012, and S1013 and stored in the memory 602. The time signature is detected (step S1014).

Then, the CPU 601 updates the value of the time signature detected in step S1014 as a new time signature value by replacing the value of the time signature variable stored in the memory 602 with the above time signature value (step S1015).

After that, the CPU 601 ends the time signature detection process of step S1001 of FIG. 10 (a) shown in the flowchart of FIG. 10 (b).

FIG. 11 is a flowchart showing a detailed example of the synchronization offset time detection process of FIG. 10A. In FIG. 11, the CPU 601 first detects the timing level for each bar time length calculated from the tempo bpm value detected by the tempo detection process in step S702 of FIG. 7, as described in step S302 of FIG. (Step S1101).

Next, the CPU 601 detects the timing at which the timing level value detected in step S1101 becomes the maximum value as the synchronization offset time, as described in step S302 of FIG. 3 (step S1102).

Then, the CPU 601 updates the maximum value detected in step S1102 as a new synchronization offset time by replacing the value of the synchronization offset time variable stored in the memory 602 with the above maximum value (step S1103).

After that, the CPU 601 ends the synchronization offset time detection process of step S1002 of FIG. 10A shown in the flowchart of FIG.

According to the above-described embodiment, tempo detection and bar and beat timing synchronization can be performed more accurately with a smaller amount of processing than, for example, a method using a Fourier transform, and fill-in, tempo synchronization effect, etc. for music can be performed in real time. Can be added.

Although the embodiments of the disclosure and their advantages have been described in detail above, those skilled in the art can make various changes, additions and omissions without departing from the scope of the present invention clearly described in the claims. ..

In addition, the present invention is not limited to the above-described embodiment, and can be variously modified at the implementation stage without departing from the gist thereof. In addition, the functions executed in the above-described embodiment may be combined as appropriate as possible. The above-described embodiments include various steps, and various inventions can be extracted by an appropriate combination according to a plurality of disclosed constitutional requirements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, if the effect is obtained, the configuration in which the constituent requirements are deleted can be extracted as an invention.

The following additional notes will be further disclosed with respect to the above embodiments.
(Appendix 1)
Acquire music waveform intensity data showing the distribution of waveform intensity from music waveform data,
The autocorrelation based on the music waveform intensity data is detected, and the basic autocorrelation value indicating the intensity for each cycle is calculated.
Specify the time signature corresponding to the song,
A calculation autocorrelation value obtained by changing the period of the basic autocorrelation value at a magnification determined based on the specified time signature is calculated.
The part where the detection autocorrelation value is maximized by weighting and adding the calculation autocorrelation value to the basic autocorrelation value is specified.
The period corresponding to the specified portion is detected as the tempo of the music.
A tempo detector with a control unit.
(Appendix 2)
In the tempo detection device according to Appendix 1,
When the designated time signature is X, the magnifications are X / (X-1) times and X / (X + 1) times.
The control unit further multiplies the period of the basic autocorrelation value by X / (X-1) to multiply the period of the basic autocorrelation value by X / (X-1), and the period of the basic autocorrelation value is multiplied by X / (X + 1). A tempo detection device that calculates the autocorrelation value for detection by weighting and adding the correlation value to the basic autocorrelation value.
(Appendix 3)
In the tempo detection device according to Appendix 1 or 2,
The control unit is a tempo detection device that further adjusts the weight value when performing the weighting addition according to the intensity of the waveform data of the music.
(Appendix 4)
In the tempo detection device according to the appendices 1 to 3,
The control unit further acquires envelope waveform data as a result of performing low-pass filter processing having a high cutoff frequency on the music waveform intensity data acquired from the waveform data of the music, and with respect to the acquired envelope waveform data. A tempo detection device that acquires a beat waveform by subtracting the result of low-pass filtering with a low cutoff frequency from the envelope waveform data, detects the autocorrelation of the beat waveform, and calculates the basic autocorrelation value. ..
(Appendix 5)
In the tempo detection device according to the appendices 1 to 4,
The control unit is a tempo detection device that further adjusts the weight value when performing the weighting addition according to the intensity of the beat waveform.
(Appendix 6)
In the tempo detection device according to the appendices 1 to 5,
The control unit further
Based on the detected tempo of the music, a plurality of time lengths including 1/3 hour length of 1 bar and 1/4 hour length of 1 bar are specified.
A plurality of timing level values are calculated by periodically synthesizing a waveform based on the differential value of the beat waveform for each of the specified time lengths.
A tempo detection device that detects the time signature corresponding to the time length that maximizes the dispersion of the timing level value as the time signature of the music.
(Appendix 7)
Acquire music waveform intensity data showing the distribution of waveform intensity from music waveform data,
The autocorrelation based on the music waveform intensity data is detected, and the basic autocorrelation value indicating the intensity for each cycle is calculated.
Specify the time signature corresponding to the song,
A calculation autocorrelation value obtained by changing the period of the basic autocorrelation value at a magnification determined based on the specified time signature is calculated.
The part where the detection autocorrelation value is maximized by weighting and adding the calculation autocorrelation value to the basic autocorrelation value is specified.
The period corresponding to the specified portion is detected as the tempo of the music.
Tempo detection method.
(Appendix 8)
Obtain the amplitude intensity envelope from the waveform data of the music and
The autocorrelation based on the envelope is detected, and the basic autocorrelation value indicating the intensity for each cycle is calculated.
Specify the time signature corresponding to the song,
A calculation autocorrelation value obtained by changing the period of the basic autocorrelation value at a magnification determined based on the specified time signature is calculated.
The part where the detection autocorrelation value is maximized by weighting and adding the calculation autocorrelation value to the basic autocorrelation value is specified.
The period corresponding to the specified portion is detected as the tempo of the music.
A program that lets a computer perform processing.

S101 Voice input processing S102 Envelope detection processing S103 Beat waveform detection processing S104 Beat intensity detection processing S105 Self-correlation detection processing S106 Tempo determination processing 201 Envelope waveform 202 Music waveform intensity 203 Beat waveform 204, 206 Original tempo value of music 205, 207 Autocorrelation Maximum value S301 Beat detection processing S302 Synchronous offset time detection processing 601 CPU
602 Memory 603 Input device 604 Output device 605 Auxiliary information storage device 606 Media drive device 607 Network connection device 608 Bus 609 Portable recording medium

Claims (8)

Acquire music waveform intensity data showing the distribution of waveform intensity from music waveform data,
The autocorrelation based on the music waveform intensity data is detected, and the basic autocorrelation value indicating the intensity for each cycle is calculated.
Specify the time signature corresponding to the song,
A calculation autocorrelation value obtained by changing the period of the basic autocorrelation value at a magnification determined based on the specified time signature is calculated.
The part where the detection autocorrelation value is maximized by weighting and adding the calculation autocorrelation value to the basic autocorrelation value is specified.
The period corresponding to the specified portion is detected as the tempo of the music.
A tempo detector with a control unit. In the tempo detection device according to claim 1,
When the designated time signature is X, the magnifications are X / (X-1) times and X / (X + 1) times.
The control unit further multiplies the period of the basic autocorrelation value by X / (X-1) to multiply the period of the basic autocorrelation value by X / (X-1), and the period of the basic autocorrelation value is multiplied by X / (X + 1). A tempo detection device that calculates the autocorrelation value for detection by weighting and adding the correlation value to the basic autocorrelation value. In the tempo detection device according to claim 1 or 2.
The control unit is a tempo detection device that further adjusts the weight value when performing the weighting addition according to the intensity of the waveform data of the music. In the tempo detection device according to claims 1 to 3,
The control unit further acquires envelope waveform data as a result of performing low-pass filter processing having a high cutoff frequency on the music waveform intensity data acquired from the waveform data of the music, and with respect to the acquired envelope waveform data. A tempo detection device that acquires a beat waveform by subtracting the result of low-pass filtering with a low cutoff frequency from the envelope waveform data, detects the autocorrelation of the beat waveform, and calculates the basic autocorrelation value. .. In the tempo detection device according to claims 1 to 4,
The control unit is a tempo detection device that further adjusts the weight value when performing the weighting addition according to the intensity of the beat waveform. In the tempo detection device according to claims 1 to 5,
The control unit further
Based on the detected tempo of the music, a plurality of time lengths including 1/3 hour length of 1 bar and 1/4 hour length of 1 bar are specified.
A plurality of timing level values are calculated by periodically synthesizing a waveform based on the differential value of the beat waveform for each of the specified time lengths.
A tempo detection device that detects the time signature corresponding to the time length that maximizes the dispersion of the timing level value as the time signature of the music. Acquire music waveform intensity data showing the distribution of waveform intensity from music waveform data,
The autocorrelation based on the music waveform intensity data is detected, and the basic autocorrelation value indicating the intensity for each cycle is calculated.
Specify the time signature corresponding to the song,
A calculation autocorrelation value obtained by changing the period of the basic autocorrelation value at a magnification determined based on the specified time signature is calculated.
The part where the detection autocorrelation value is maximized by weighting and adding the calculation autocorrelation value to the basic autocorrelation value is specified.
The period corresponding to the specified portion is detected as the tempo of the music.
Tempo detection method. Obtain the amplitude intensity envelope from the waveform data of the music and
The autocorrelation based on the envelope is detected, and the basic autocorrelation value indicating the intensity for each cycle is calculated.
Specify the time signature corresponding to the song,
A calculation autocorrelation value obtained by changing the period of the basic autocorrelation value at a magnification determined based on the specified time signature is calculated.
The part where the detection autocorrelation value is maximized by weighting and adding the calculation autocorrelation value to the basic autocorrelation value is specified.
The period corresponding to the specified portion is detected as the tempo of the music.
A program that lets a computer perform processing.

Images