KR20190086872A - Method the create midi file with harmonics removed and method to converse audio file - Google Patents

Abstract

The disclosed technology relates to a method of creating a MIDI file having harmonics removed and a method of converting an audio file. The method of creating a MIDI file having harmonics removed comprises: a step in which a system receives an audio file and analyzes a frequency of the audio file; a step in which the system removes harmonics by setting an upper threshold value according to a level of the frequency; a step in which the system sets a window having a different size for each frequency region corresponding to each sound included in the frequency from which the harmonics are removed; a step which obtains a representative value for each window by applying the weight to each of the windows having different sizes, and calculates starting and ending time points for the representative value; and a step which converts the audio file into a MIDI file by selecting a MIDI sound matching the representative value on the basis of the starting and ending time points. Accordingly, the MIDI file having accurate pitch and beats as compared with the original audio file can be created, and the audio file can be accurately converted into a sound file in a different format.

Description

METHOD AND METHOD TO CONVERSE AUDIO FILE BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method of generating a MIDI file,

The disclosed technique relates to a method of generating a more accurate MIDI file by removing harmonics when converting an audio file into a MIDI file, and a method of converting an audio file into a file of a different format.

The MIDI file conversion method extracts frequency values corresponding to MIDI sounds and converts the files through a series of matching operations. Generally, when an audio file is input to the system and the system extracts frequency values for the audio file, a MIDI file is generated based on the frequency value.

However, in extracting the frequency value for the audio file, harmonic frequencies due to harmonics are mixed. That is, in the process of converting to a MIDI file, a harmonic frequency is recognized as a sound and converted to a MIDI sound, so that a slightly different result than an original audio file is generated or an incorrect MIDI file is generated.

US Patent Application Publication No. 2006-0075884 discloses a technique for extracting a melody by scaling a signal of a recognizable region in a frequency domain with reference to a method and device for extracting a melody underlying an audio signal. However, since harmonics are not removed prior to MIDI conversion, there is a disadvantage in that an incorrect or somewhat different MIDI file is generated as in the prior art.

The disclosed technique is to provide a method of generating a more accurate MIDI file by removing harmonics at a frequency of an audio file and applying windows of different sizes to each area, and a method of converting an audio file into a sound source file of a different format.

According to a first aspect of the present invention, there is provided a method of receiving a sound file and analyzing a frequency of the audio file, the method comprising: setting an upper threshold value according to a level of the frequency to remove harmonics; Setting a window of a different size for each frequency region corresponding to each sound included in a frequency at which the system eliminates the harmonics, applying a weight to each of the windows of different sizes, Calculating a start point and an end point of the representative value, and selecting a MIDI note matching the representative value based on the start point and the end point to convert the audio file into a MIDI file, Generating method.

According to a second aspect of the present invention, there is provided a method for receiving an audio file and analyzing a frequency of the audio file, the system comprising the steps of setting an upper threshold value according to a level of the frequency, Setting a window of a different size for each frequency region corresponding to each sound included in a frequency at which the system eliminates the harmonics, applying a weight to each of the windows of different sizes, Calculating a starting point and an ending point of the representative value and calculating a pitch and a beat matched to the representative value based on the starting point and the end point to calculate an audio file from which the harmonic is removed into a sound source file of a different format An audio file conversion method comprising a step of converting To provide.

Embodiments of the disclosed technique may have effects that include the following advantages. It should be understood, however, that the scope of the disclosed technology is not to be construed as limited thereby, since the embodiments of the disclosed technology are not meant to include all such embodiments.

According to an embodiment of the disclosed technology, a harmonic-free MIDI file generation method has the effect of generating a MIDI file with high accuracy by removing harmonics when converting a MIDI file.

Further, there is an advantage that the pitch and beat of a specific sound can be calculated more accurately by applying windows of different sizes to each frequency region.

1 is a flowchart illustrating a method of generating a harmonized MIDI file according to an embodiment of the disclosed technique.
2 is a flowchart of a conventional method of generating a MIDI file.
3 is a diagram illustrating removing harmonics according to one embodiment of the disclosed technique.
Figure 4 illustrates creating windows of different sizes according to one embodiment of the disclosed technique.
5 is a diagram illustrating calculation of a representative value according to one embodiment of the disclosed technique.
6 is a flowchart of an audio file conversion method according to an embodiment of the disclosed technique.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, A, B, etc., may be used to describe various components, but the components are not limited by the terms, but may be used to distinguish one component from another . For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

It is to be understood that the singular < RTI ID = 0.0 > terms < / RTI > used herein should be interpreted to include a plurality of representations unless the context clearly dictates otherwise. And "comprises ", when used in this specification, specify the presence of stated features, numbers, steps, operations, elements, parts, or combinations thereof, Or combinations thereof, as a matter of course.

Before describing the drawings in detail, it is to be clarified that the division of constituent parts in this specification is merely a division by main functions of each constituent part. That is, two or more constituent parts to be described below may be combined into one constituent part, or one constituent part may be divided into two or more functions according to functions that are more subdivided.

In addition, each of the constituent units described below may additionally perform some or all of the functions of other constituent units in addition to the main functions of the constituent units themselves, and that some of the main functions, And may be carried out in a dedicated manner. Accordingly, the presence or absence of each component described in this specification should be interpreted as a function.

1 is a flowchart illustrating a method of generating a harmonized MIDI file according to an embodiment of the disclosed technique. Referring to FIG. 1, a method of generating a MIDI file with harmonics removed includes the following steps.

In step 110, the system analyzes the frequency by receiving the audio file. The system includes a processor having a memory capable of storing audio files and MIDI files, and performing an operation of analyzing frequencies of the audio files and converting the analyzed frequencies into MIDI files. For example, the system may be a terminal such as a computer or a notebook.

On the other hand, when the system receives an audio file in the memory, it analyzes the frequency of the audio file using the processor. In one embodiment, the fundamental frequency and harmonics for each note included in the audio file can be analyzed. Here, the fundamental frequency is a frequency with respect to the original sound, which means a frequency including components such as different pitches, beats, and pitches for each note. Harmonics are frequencies for harmonics generated by superposition of sounds, which are n times the fundamental frequency.

Referring to FIG. 2, if a wave file or an MP3 file is input to the system according to step 210 in the conventional method of generating a MIDI file, the frequency of the corresponding file is analyzed to match the frequency of the MIDI sound. In step 230, a MIDI file is converted.

That is, since a simple process of extracting a frequency of an input audio file and converting it to a MIDI file is performed, the accuracy of the pitch and beat of each sound is degraded, and the MIDI file conversion process is performed without removing harmonics Therefore, there is a problem that a MIDI file is converted into a MIDI sound until an unnecessary sound is generated, thereby generating a somewhat inaccurate MIDI file as compared with the original audio.

Accordingly, in order to solve such a problem, the disclosed technology divides the fundamental frequency and the harmonic included in the audio file to remove the harmonics, and then generates the MIDI file using only the remaining fundamental frequencies.

In step 120, the system sets an upper threshold according to the level of the frequency to remove the harmonics. In one embodiment, if there is a larger frequency at the same time as a frequency that satisfies a certain level, it can be determined as a harmonic. Generally, since the harmonics appear at frequencies n times larger than the fundamental frequency at the same time, it is possible to remove the harmonics by removing the large values by dividing the frequency by the viewpoint.

In another embodiment, a frequency is classified into a plurality of regions corresponding to one sound, and when two frequencies are detected for each region, a large frequency may be determined as a harmonic and removed. Referring to FIG. 3, it can be seen that frequencies extracted from an audio file are classified into a plurality of regions according to a viewpoint, and when two frequencies are detected for each region, a large frequency 301 is determined as a harmonic.

Here, the sound included in the audio file has a relatively large sound and a small sound, but it is possible to distinguish the two according to the difference between the large and small sound. For example, as shown in FIG. 3, the system can remove the harmonics 301 because the level of the harmonics 301 is higher than the fundamental frequency 302 in the second region.

In step 130, the system sets a window having a different size for each frequency region corresponding to each sound included in the remaining frequency from which the harmonics are removed. Here, the window is not a reference for dividing a sound according to time, but refers to a section or an area which is set to a different size for each sound by applying different threshold values to the respective frequency regions corresponding to the respective sounds.

4, frequencies for three notes are displayed, and five windows 401, 402, 403, 404, and 405 of different sizes for each note are applied to the threshold, As shown in FIG. Here, the size of the window is defined as a time when the slope of the frequency changes from 0 to a positive value based on a threshold value applied to a frequency region corresponding to one note, and a moment when the value changes from a negative value to 0 as a starting point and an ending point of the window for the corresponding frequency region Setting.

That is, the window is set not only to interfere with harmonics but also to distinguish clearly from other sounds and to judge each sound more accurately. It is possible to detect similar sounds through windows if several sounds of similar sounds are present at the same time. If a large sound and a small sound are generated at the same time, unlike the former, it is very difficult to detect it, so that the sound of the third window 403 is discarded.

Even if you hear the original audio file as it is, you can not recognize it easily because a small sound is buried when a big sound and a small sound appear at the same time. If you convert it to MIDI, even if you delete a small sound, Results are derived. Therefore, it is possible to select only the optimum frequency for generating a MIDI file by applying windows of different sizes.

In step 140, the system obtains a representative value for each window by applying a weight to each window having a different size. Then, the start and end points of the representative value are calculated. In one embodiment, the system may calculate a frequency value corresponding to a highest level of a corresponding window to calculate a pitch and a beat for a note included in one window, using a representative value.

Then, the start and end points of each representative value are calculated and the pitch and beat of the corresponding note are calculated. For example, referring to FIG. 5, a point corresponding to the highest level of a frequency corresponding to one note is set as the negative representative value 501, and the starting point 502 and the ending point 503). The time from the start point 502 to the end point 503 can be calculated as a time for expressing the MIDI sound corresponding to the representative value 501 and a duration for continuing the MIDI sound. Through this process, it is possible to calculate how long each sound is to be sounded and how much time it takes to convert each sound included in the audio file into MIDI sound. Therefore, it is possible to generate a MIDI file having a more accurate pitch and beat than the conventional MIDI conversion method.

In step 150, the system selects a MIDI sound that matches the representative value based on the start point and the end point of the representative value obtained in step 140. In one embodiment, the system stores a table for MIDI notes in advance, and may generate MIDI files in a manner that matches each of the MIDI notes that match each note of the audio file converted to frequency. Although the method of generating a MIDI file itself is not significantly different from the conventional technique, when a MIDI file is generated according to the disclosed technology, unnecessary sounds are not added to the MIDI file because frequencies using no harmonics are used Since the pitch and the beat of each note are more accurately determined by applying the window, it is advantageous to generate a MIDI file that expresses each note as a simple note in a more detailed manner.

Meanwhile, referring to FIG. 6, the system can convert an audio file into a sound file of a different format as well as a MIDI file. For example, the system can perform steps 610 to 640 in the same manner as steps 110 to 140, thereby obtaining the harmonics free from the original audio file. In step 650, the pitch and the pitch can be calculated using the dynamic window set to the frequency corresponding to each note.

That is, since the pure frequencies of the tones included in the audio can be obtained, it can be converted into various digital sound source formats such as WAVE, FLAC, OGG, and AAC based on this frequency. Of course, in order to convert sound source files of different formats, the system must store the data according to the format in advance.

Although the harmonic-free MIDI file generation method and the audio file conversion method according to an embodiment of the disclosed technology have been described with reference to the embodiments shown in the drawings to facilitate understanding, the present invention is not limited thereto. Those skilled in the art will appreciate that various modifications and equivalent embodiments are possible. Accordingly, the true scope of protection of the disclosed technology should be determined by the appended claims.

110: Frequency analysis of audio files 120: Harmonic elimination
130: Window setting for each frequency domain 140:
150: Convert to MIDI file

Claims (10)

The system receiving an audio file and analyzing the frequency for the audio file;
The system setting an upper threshold according to the level of the frequency to remove the harmonics;
Setting a window of a different size for each frequency region corresponding to each sound included in a frequency at which the system eliminates the harmonics;
Calculating a representative value for each window by applying a weight to each window of the different size, and calculating a starting point and an end point of the representative value; And
And converting the audio file into a MIDI file by selecting a MIDI sound matching the representative value based on the start point and the end point. The method of claim 1, wherein analyzing the frequency comprises:
And analyzing harmonics of harmonics generated by overlapping the fundamental frequencies and the sounds of the respective notes included in the audio file. 2. The method of claim 1, wherein removing the harmonic comprises:
And if a larger frequency exists at the same time as the frequency satisfying the level, it is determined to be a harmonic and removed. 2. The method of claim 1, wherein removing the harmonic comprises:
The method comprising the steps of: classifying the frequency into a plurality of regions corresponding to one sound, and when two frequencies are detected for each region, determining a higher frequency as a harmonic and removing the higher frequency; 2. The method of claim 1, wherein the setting of the window comprises:
And setting different windows of different sizes by applying different threshold values to respective frequency regions corresponding to the respective tones. 2. The method of claim 1, wherein the setting of the window comprises:
A threshold value is applied to a frequency region corresponding to one note, and a moment when the slope of the frequency changes from 0 to a positive value on the basis of the threshold value and a moment that changes from a negative value to 0 is defined as a start point and an end point of the window for the frequency region How to create a MIDI file to set. The method according to claim 1,
Wherein the representative value is calculated as the representative value of the frequency corresponding to the highest level of the window in order to calculate a pitch and a beat for a note included in one window. 2. The method of claim 1, wherein the calculating the start time and end time comprises:
And calculating the time from the start point to the end point as a time for expressing the MIDI sound corresponding to the representative value and a duration for continuing the MIDI sound. 2. The method of claim 1,
A method for generating a MIDI file by mapping a MIDI note matched to the representative value using a pre-stored MIDI table and converting the MIDI note to a file. The system receiving an audio file and analyzing the frequency for the audio file;
The system setting an upper threshold according to the level of the frequency to remove the harmonics;
Setting a window of a different size for each frequency region corresponding to each sound included in a frequency at which the system eliminates the harmonics;
Calculating a representative value for each window by applying a weight to each window of the different size, and calculating a starting point and an end point of the representative value; And
Calculating a pitch and a beat matched to the representative value based on the start point and the end point, and converting the harmonics-free audio file into sound source files in different formats.

Images