JP2007156187A - Music processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To calculate the music information showing the types composing the music and the sections of those various types even when the music includes key changes. <P>SOLUTION: The tone scale information input section 11 accepts the input tone scale information of a music. The appearance probability calculator 12 calculates the appearance probability of the tone included in the tone scale information from the tone scale information input to the tone scale information input section 11. The weight calculator 14 calculates the weight of each template from the appearance probability distribution of the calculated music and on the templates different for the types of the music stored in the template memory 13. The music information calculator 15 calculates the music information showing the types composing the music and the sections of those types from the calculated weight. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a music processing device, and more specifically, a music processing device that calculates music information representing each type constituting the music and a section in which each type exists from the scale information of the music, and a calculation The present invention relates to a music processing apparatus that searches for music using the music information that has been performed.

  Conventionally, methods for detecting the type of music are known. The type of music indicates an element (for example, key) that determines the image of the music. For example, as a method for detecting the key of a music piece, the appearance probability of each pitch name included in the scale information is calculated based on information about the musical scale (hereinafter referred to as scale information), and the appearance probability of each pitch name is calculated. A method of detecting the key of a music using a distribution (referred to as an appearance probability distribution; see FIG. 3 described later) is known. In this method, an ideal appearance probability distribution created in advance is prepared as a template for each of a plurality of types of keys. And the appearance probability distribution of the music used as the object which detects a key is calculated, and the appearance probability distribution about the said music and each said template are collated one by one. As a result, the key indicated by the template whose appearance probability distribution is closest to the music is determined as the key of the music.

  Here, the actual music includes music composed of a plurality of types of keys including modulation. The appearance probability distribution of the music including such a modulation may be a distribution different from any of the appearance probability distributions of each key included in the music as a result of combining the appearance probability distributions of a plurality of types of keys. . In the above-described conventional method, the appearance probability distribution of the music and the appearance probability distribution of the template are collated on a one-to-one basis. Therefore, when the appearance probability distribution of the music is a composite distribution of a plurality of keys, the template that most closely approximates the composite distribution. The key indicated by is determined as the key of the music. That is, in this case, a key different from any of the keys included in the music is determined as the key of the music. Therefore, in the above conventional method, there is a possibility that a key is erroneously detected for music composed of a plurality of types of keys including a modulation. In the above-described conventional method, the appearance probability distribution of the music and the appearance probability distribution of the template are collated on a one-to-one basis, so that only the key of the entire music can be detected. It was not possible to calculate music information indicating the section where each key exists.

Therefore, conventionally, a method has been proposed in which input scale information is divided into predetermined sections and the key of each section is detected (see, for example, Patent Document 1). In this method, for example, musical scale information of music is divided into a plurality of sections such that the first bar to the fourth bar of the music are one section, and the second bar to the fifth bar are the next section. Then, the key of the section is detected for each section. In other words, music information representing each key constituting the music and a section where each key exists is calculated. Furthermore, the part to be modulated is detected by the detected change in the tone of each section.
Japanese Patent No. 2715816

  Here, in the method disclosed in Patent Document 1, the target whose key is to be detected is not the entire input scale information, but a section obtained by dividing the scale information into short sections. Specifically, in the method of Patent Document 1, in order to detect modulation, scale information must be divided into units of at least about several measures (for example, about four measures). However, if the scale information is divided into sections, the number of sounds included in the key detection target is reduced. That is, by dividing the scale information into sections, the number of pieces of information included in the key detection target is reduced, so that the accuracy of key detection is inevitably lowered. As described above, in the method of Patent Document 1, the accuracy of detecting the key of each section is lowered because the scale information must be divided, and the music information of the music cannot be accurately calculated. .

  Therefore, the present invention provides a music processing apparatus capable of accurately calculating music information that represents each type constituting a music and a section in which each type exists even if the music includes modulation. The purpose is to provide.

  1st invention is a music processing apparatus which calculates the music information showing each type which comprises the said music part, and the area where each said type exists about the music part which is at least a part of predetermined music, The appearance probability acquisition unit that acquires the distribution of the appearance probability for each musical name for the music part, and the template that is different for each type of music and that shows the distribution of the appearance probability for each musical name for each of the plurality of types A template storage unit, and for each of a plurality of templates stored in the template storage unit, the load degree indicating the ratio of the template in the distribution of the appearance probability of each pitch name acquired by the appearance probability acquisition unit The total length of the sections in which the load degree calculation unit for calculating the load degree for each and the types constituting the music part exist are occupied in the entire music part. As the ratio and the size of the type of load degree are matched respectively for each type that includes a music information calculating unit that calculates the music information.

  In a second aspect based on the first aspect, the music information calculation unit, for each type constituting the music part, determines the length of the section in which each type exists, and the magnitude of the load degree of each type. A part of the music part so that the setting unit for setting the temporary music information having the matching length and the type constituting the music part and the type represented by the temporary music information match in each section represented by the temporary music information. Is replaced in the music part, a replacement information storage unit that stores replacement information indicating contents replaced in the replacement unit, and a replacement order in the replacement unit based on replacement information stored in the replacement information storage unit. And a reverse replacement unit that replaces part of the provisional music information in the reverse order.

  In a third aspect based on the second aspect, the replacement unit calculates the degree of difference between the type represented by the provisional music information and the type constituting the musical piece part in each section represented by the provisional music information, The process of replacing a part of the music part in the music part so as to reduce the degree of difference is repeated, and when the degree of difference becomes smaller than a predetermined value, the process ends, and the replacement information storage unit Each time the section performs a process of replacing a part of the music part, replacement information is stored.

  According to a fourth aspect of the present invention, in the first aspect, the appearance probability acquisition unit is calculated by a music data storage unit that stores data of a plurality of pieces of music for which an appearance probability distribution is acquired, and a music information calculation unit. The music information is stored in the music data storage unit in association with the music corresponding to the music information among the music data stored in the music data storage unit, and the music data storage unit stores the data. And a search unit that searches the stored music using the music information.

  According to a fifth aspect, in the fourth aspect, the apparatus further includes a designated part input unit that receives an input for designating a part of the song by the user, and the search unit stores the song information stored in the song data storage unit. Among these, a search is performed using a part of music information designated by the designated part input unit.

  A sixth invention further includes a music selection rule storage unit that stores a music selection rule that associates music selection information to be input by a user with a condition related to music information in the fourth invention, and the search unit selects the music selection by the user When the information is input, the music information associated with the music selection information associated with the inputted music selection information is detected among the music information associated with the music whose data is stored in the music data storage unit, The music associated with the detected music information is output as a search result.

  7th invention is used in the music processing apparatus which calculates the music information showing each type which comprises the said music part, and the area where each said type exists about the music part which is at least a part of predetermined music. In the music processing apparatus, templates that are different for each type of music and that indicate the distribution of appearance probabilities for each pitch name are stored in advance for each of the plurality of types. Appearance probability acquisition step of acquiring the appearance probability distribution for each, and for each of a plurality of templates stored in the music processing device, of the template occupying the distribution of the appearance probability of each pitch name acquired in the appearance probability acquisition step There is a load degree calculation step for calculating the load degree for each type, and a type constituting the music part. As the total length of the interval and the magnitude of the ratio and the type of load level in the whole of the music piece portions match each for each type, and a music information calculation step of calculating the music information.

  In an eighth aspect based on the seventh aspect, the music processing apparatus further stores in advance data of a plurality of pieces of music from which the distribution of appearance probabilities is acquired in the appearance probability acquisition step. A music information storage step for storing the music information calculated in the calculation step in the music processing apparatus in association with the music corresponding to the music information among the music data stored in the music processing apparatus; And a search step of searching for music in which data is stored using music information.

  According to a ninth aspect of the present invention, there is provided a computer of a music processing apparatus that calculates music information representing each type constituting the music part and a section in which each type exists for a music part that is at least a part of the predetermined music. In the music processing apparatus, a template that is different for each type of music and indicates a distribution of appearance probabilities for each pitch name is stored in advance for each of a plurality of types. For the appearance probability acquisition step of acquiring the appearance probability distribution for each pitch name, and for each of the plurality of templates stored in the music processing device, the distribution of the appearance probability of each pitch name acquired in the appearance probability acquisition step A load degree calculation step for calculating a load degree for each type, which is a load degree indicating a ratio of the template to be occupied, and a music part. A music information calculation step for calculating music information so that the total length of the section in which the type to be present occupies in the music part and the magnitude of the load degree of the type are the same for each type. A program to be executed by a computer.

  A tenth invention is a computer-readable recording medium on which the program according to the ninth invention is recorded.

  An eleventh aspect of the invention is an integrated circuit that calculates music information representing each type constituting the music part and a section in which each type exists for a music part that is at least a part of the predetermined music, Appearance probability acquisition unit that acquires distribution of appearance probabilities for each musical name for music parts, and acquisition of appearance probabilities for a plurality of templates that are different for each type of music and that show the distribution of appearance probabilities for each musical name The degree of load indicating the ratio of the template in the distribution of the appearance probability of each pitch name acquired by the part, and a load degree calculating part for calculating the load degree for each type, and the section in which the type constituting the music part exists The music information calculation is performed to calculate the music information so that the ratio of the total length of the music portion to the entire music portion and the magnitude of the load degree of the type are the same for each type. And a part.

  According to the first aspect of the present invention, instead of selecting one template that most closely matches from a plurality of templates, the degree of load of each template is calculated. That is, instead of determining one type of music corresponding to the template for a music part that is at least a part of a predetermined music (music for which music information is to be calculated), "Percentage" is calculated. Therefore, for example, when the music part is composed of two tones, the degree of load of the templates of the two tones is calculated to be large. That is, even if the music includes modulation, each type constituting the music can be accurately calculated using the load degree. Then, using the degree of load obtained by accurately calculating each type constituting the music part, the music information is transposed by calculating music information representing each type constituting the music part and a section in which each type exists. Even if it contains, it becomes possible to calculate correctly the music information of the said music. In addition, by calculating the music information, it is possible to present to the user a section in which each type constituting the music part exists, and the user can accurately determine in which section each type constituting the music part exists. Can grasp.

  According to the second aspect of the present invention, a part of the music part is included in the music part so that the type constituting the music part and the type represented by the temporary music information match in each section represented by the temporary music information. The music information of the music portion can be accurately calculated by exchanging a part of the provisional music information in the reverse order of the exchange order based on the exchange information after the exchange.

  According to the third aspect, the replacement information stored in the replacement information storage unit is replacement information that is replaced so that the degree of difference is small, and the replacement information is used to efficiently calculate music information. can do.

  According to the fourth aspect, music can be searched using accurate music information, and high-precision search can be realized.

  According to the fifth aspect, it is possible to perform a search for a part of the designated music piece and to realize a more flexible search.

  According to the sixth aspect, the user can easily search for music related to the input music selection condition.

  Embodiments of the present invention will be described below with reference to the drawings. In the following description, “music information” is information representing each type constituting a music and a section in which each type exists. Specifically, the music information is information represented by a combination of information on a section in the music and the type of the section. The “type” is an element that determines the image of the music, and indicates, for example, tonality, scale, and tone. Here, “tonality” means two types of major and minor. “Scale” means 12 major and minor combinations with the same key signature. “Tone” means 24 types (C: C major, Am: A minor, etc.) composed of tonality and scale. As an example of a music processing apparatus according to the present invention, a music information calculation apparatus that calculates the music information will be described as a first embodiment. As another example of the music processing apparatus according to the present invention, a music search apparatus 2 that searches for music using the music information will be described as a second embodiment.

(First embodiment)
First, the music information calculation apparatus 1 according to the first embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing a music information calculation apparatus 1 according to the first embodiment of the present invention. In FIG. 1, the music information calculation device 1 includes a scale information input unit 11, an appearance probability calculation unit 12, a template storage unit 13, a load degree calculation unit 14, a music information calculation unit 15, and a replacement information storage unit 16.

  The scale information input unit 11 accepts input of musical scale information from outside or inside the music information calculation apparatus 1. The musical scale information input to the musical scale information input unit 11 is data including at least information on the pitch (note number) and the length (duration) of the sound. FIG. 2 is a diagram illustrating an example of musical scale information input to the musical scale information input unit 11. In FIG. 2, the musical scale information is expressed by the time indicated by “number of bars / number of beats / number of ticks”, velocity representing the intensity of sound, and note number. Here, the time indicates the time when each note number is sounded or muted. The velocity is expressed by an integer in the range of 0 to 127, and the volume increases as the velocity value increases. A velocity value of “0” indicates mute. The value of the note number is set with “60” being a value indicating the height of C at the center of the piano keyboard. In FIG. 2, the duration is represented by time and velocity. For example, when the time is “000/0/000”, the note number is “60” and the velocity is “90 (pronunciation)”. When the time is “000/0/432”, the note number is “60” and the velocity is “0 (silence)”. Therefore, the scale information in the time interval from “000/0/000” to “000/0/432” has a note number of “60”, a velocity of “90”, and a duration of 432 ticks.

  In the example of the musical scale information shown in FIG. 2, the time is indicated by “measure number / beat number / tick number”, but may be indicated by “hour / minute / second / frame / subframe”, for example. Good. The velocity is expressed by an integer in the range of 0 to 127. For example, the velocity may be expressed by a binary value of “1” indicating sound generation and “0” indicating mute.

  As described above, the musical scale information shown in FIG. 2 includes information on the pitch (note number) and the length (duration) of the sound. Note that the musical scale information may simply be expressed by a method having a one-to-one duration with respect to a note number. In addition to the note number and duration, the musical scale information may be expressed by a method that further includes pitch bend information representing a continuous change in the note number. At this time, the note number varies according to the pitch bend value.

  The appearance probability calculation unit 12 calculates the appearance probability of each pitch name included in the musical scale information input to the musical scale information input unit 11. FIG. 3 is a diagram conceptually showing the distribution of appearance probabilities (appearance probability distribution) for each pitch name. In FIG. 3, the pitch name numbers 0 to 11 are assigned to the 12 types of pitch names, such as C; i = 0, C #; i = 1,..., B; Yes. In the following description, a pitch name having a pitch name number i may be referred to as “pitch name i”.

  The template storage unit 13 stores 24 types of templates corresponding to 24 types of keys. The 24 types of templates represent different types of music. The template also shows an ideal appearance probability distribution of the corresponding key. Each of these templates is created in advance and stored in the template storage unit 13. FIG. 4 is a diagram illustrating an example of 24 types of templates and calculation results of the degree of load for each template. In FIG. 4, key numbers j are assigned to 24 types of keys indicated by each template, with the major key j = 0-11 and the minor key j = 12-23. In the following description, a key whose key number is j may be referred to as “key j”. The calculation result of the degree of load for each template will be described later.

  The load degree calculation unit 14 calculates the load degree for each template based on the appearance probability distribution of the music calculated by the appearance probability calculation unit 12 and each template stored in the template storage unit 13 (FIG. 4). reference.). The load degree calculation unit 14 calculates 24 load degrees so as to correspond to 24 types of templates one by one. Here, the degree of load indicates the ratio (occupancy) of the appearance probability distribution of the template in the appearance probability distribution of music. In other words, the degree of template load is a value indicating the contribution ratio of the appearance probability distribution of a template when the appearance probability distribution of the music is expressed using 24 types of templates. For example, a template having a large value of the degree of load is a template having a large proportion of the appearance probability distribution of music. Therefore, the key corresponding to the template with a large value of the degree of load is a key that occupies a large proportion of the music.

  The music information calculation unit 15 is a load indicating the ratio of the total length of the sections in which the key constituting the music exists to the entire music and the ratio of the key template calculated by the load degree calculation unit 14. The music information of the music is calculated so that the magnitude of the degree matches for each key. Detailed processing for calculating music information will be described later. In the replacement information storage unit 16, replacement information of sections generated in the process of the music information calculation unit 15 is stored.

  Next, a processing flow of the music information calculation apparatus 1 according to the first embodiment will be described with reference to FIG. FIG. 5 is a flowchart illustrating a process flow of the music information calculation apparatus 1 according to the first embodiment.

  The musical scale information input unit 11 receives input of musical scale information from outside or inside the music information calculation apparatus 1 (step S1). For example, scale data indicating scale information, such as SMF (standard MIDI file), is input. Note that the scale data to be input may be converted from voice data such as PCM data. Here, it is assumed that the scale information as shown in FIG. 2 is input. In the present invention, since it is not necessary to divide the input music into sections such as measures, the scale information used in the present invention does not need to include information indicating the position of the measure. In the present invention, the key included in the music can be accurately detected without dividing the music including the modulation into sections such as measures.

After step S1, the appearance probability calculation unit 12 calculates the appearance probability of each pitch name included in the scale information of the entire music input in step S1 (step S2). Here, let P (i) be the appearance probability of the pitch name i. The appearance probability P (i) is calculated by dividing the sum of durations of the pitch name i included in the scale information by the sum of durations of all the pitch names (i = 0 to 11). The total sum ΣP (i) of the appearance probabilities of each pitch name is expressed as in the following formula (1).
ΣP (i) = 1 (i = 0 to 11) (1)

  As shown in step S <b> 2, in this embodiment, the appearance probability distribution of each pitch name of the music is calculated by the appearance probability calculation unit 12. Here, when the appearance probability distribution of each sound name of the music is known in advance, that is, when data indicating the appearance probability distribution of each sound name of the music is obtained, the appearance probability calculating unit 12 performs the step The data may be acquired in S2. In this case, the process of calculating the appearance probability distribution in step S2 can be omitted.

  After step S2, the load degree calculation unit 14 calculates the load degree for each template (step S3). The degree of load is calculated using the appearance probability (actual appearance probability of music) calculated in step S2 and the appearance probabilities indicated by the 24 types of templates stored in the template storage unit 13. . Hereinafter, a method for calculating the degree of load will be described in detail.

First, let us consider expressing the appearance probability distribution of music for which music information is calculated using the appearance probability distributions of 24 types of templates corresponding to 24 types of keys (j = 0 to 23). Here, the load degree indicating the ratio of the appearance probability distribution of the key j template in the appearance probability distribution of the target music is defined as W (j). Then, the appearance probability of the pitch name i included in the template of key j is Pt (j, i). At this time, the following expressions (2) and (3) are satisfied.
ΣW (j) = 1 (j = 0-23) (2)
ΣPt (j, i) = 1 (i = 0 to 11) (3)
Here, using each template (that is, using W (j) and Pt (j, i)), the appearance probability Pf (i) of the pitch name i included in the scale information of the target music is expressed. In this case, the appearance probability Pf (i) is expressed by the following equation (4).
Pf (i) = Σ (W (j) * Pt (j, i)) (j = 0 to 23) (4)
Therefore, the load degree W (j) (j) of each template is set so that the actual appearance probabilities P (i) and Pf (i) calculated in step S2 are the same for all pitch names (i = 0 to 11). = 0 to 23). That is, W (j) that satisfies the relations of equations (2) to (4) and satisfies P (i) = Pf (i) for i = 0 to 11 may be obtained. Specifically, here, by minimizing the sum of squares of errors (P (i) −Pf (i)) of the appearance probabilities of the pitch names, the load degree W (j) (j = 0-23). More specifically, let E (i) be the error of the appearance probability of the pitch name i, and let E be the sum of the squares of the error E (i), then E (i) and 'E can be expressed by Equation (5), respectively. ) And equation (6).
E (i) = P (i) -Pf (i) (5)
'E = Σ (E (i)) ² (i = 0 to 11) (6)
Using this equation (6), the degree of load W (j) (j = 0 to 23) of each template that minimizes 'E' is calculated. Note that the load degree W (j) (j = 0 to 23) of each template can be calculated by using, for example, an evolutionary strategy (Evolutionary Strategies). What algorithm is used to calculate W (j)? It may be. As described above, in step S3, the load degree W (j) (j = 0 to 23) of each template is calculated. The load degree W (j) (j = 0 to 23) of each template is a calculation result as shown in FIG. 4, for example.

  Next to step S3, the music information calculation unit 15 determines the ratio of the total length of the sections in which the key constituting the music is present to the entire music and the ratio of the key template calculated in step S3. The music information of the music is calculated so that the magnitude of the load degree indicating the same corresponds to each tone (step S4). As described above, the music information is information represented by a combination of the information of the section in the music and the key of the section. For example, [start position of the section−end position of the section: key j of the section] It is the information that is represented. For example, suppose that the entire music to be calculated is 100% (1.0), the first 60% (0.6) is E major, and the second 40% (0.4) is F major. At this time, the first half section where the start position is 0% (0.0) where the music starts and the end position is 0% to 60% (0.6) where the music starts is E major (j = 4). Further, the latter half section having the 60% position as the start position and the position where the music ends 100% (1.0) is the end position is F major (j = 5). Therefore, the music information of this music is expressed by ([0.0-0.6: 4], [0.6-1.0: 5]).

Hereinafter, the process of calculating music information will be described in detail with reference to FIG. FIG. 6 is a flowchart showing a flow of processing for calculating music information. In addition, as a premise for explaining the process of calculating the music information, the music from the start 0% (0.0) position to the 20% (0.2) position of the music is Cm, The section from the 20% position to the 50% (0.5) position is D, the section from the 50% position to the 60% (0.6) position is Cm, and from the 60% position to the music It is assumed that a song whose section to the end 100% (1.0) is D is D. That is, as shown in the uppermost part of FIG. 7, the actual music to be calculated is ([0.0-0.2: 15], [0.2-0.5: 2], [0.5- 0.6: 15], [0.6-1.0: 2]). However, the key number of Cm is j = 15, and the key number of D is j = 2. FIG. 7 is a diagram schematically showing data of each piece of music in a part of the process for calculating the music information. In this case, the load degree W (j) calculated in step S3 is a value shown in Expression (7).
W (2) = 0.7, W (15) = 0.3 (7)
According to the calculation result of the degree of load, it can be seen that the key constituting the music to be calculated includes only the key of D and Cm. It can also be seen that the total length of each section in the D key is equivalent to 70% of the entire music. It can also be seen that the total length of each section in the Cm key is equivalent to 30% of the entire music.

  On the basis of the load degree calculation result shown in Equation (7), the music information calculation unit 15 sets the music information of the music to be calculated, for example, ([0.0 -0.7: 2], [0.7-1.0: 15]) (step S401). Hereinafter, this assumed music information is referred to as provisional music information.

  After step S401, the appearance probability calculation unit 12 calculates the appearance probability of each pitch name for each section (hereinafter referred to as a temporary section) represented by temporary music information for the music to be calculated input to the scale information input unit 11. Calculate (step S402). That is, the appearance probability calculation unit 12 includes the appearance probability Pa (i) of the pitch name i included in the scale information of the first half section [0.0-0.7] for the music to be calculated, and the second half section [0. 7-1.0] and the appearance probability Pb (i) of the pitch name i are respectively calculated.

After step S402, the music information calculation unit 15 generates the appearance probability (Pa (i) and Pb (i)) of each pitch name for each temporary section of the music to be calculated, and each temporary music information represented by the temporary music information. An error function for the entire music is calculated using the appearance probabilities of each pitch name included in the template of the section key (D and Cm) (step S403). Here, the appearance probability Pa (i) of the pitch name i in the first half section in the music to be calculated, and the appearance probability Pt of the pitch name i included in the template of the key D (j = 2) assumed as the key of the first half section. Let Ea (i) be the error from (2, i). In addition, the appearance probability Pb (i) of the pitch name i in the latter half section in the song to be calculated and the appearance probability Pt () of the pitch name i included in the template of the key Cm (j = 15) assumed as the key of the latter half section. 15, i) is Eb (i). Further, the sum of squared Ea (i) is set as 'Ea', and the sum of squared Eb (i) is set as 'Eb'. At this time, Ea (i), 'Ea, Eb (i), and' Eb are expressed by Expression (8) to Expression (11), respectively.
Ea (i) = Pa (i) −Pt (2, i) (8)
'Ea = Σ (Ea (i)) ² (i = 0 to 11) (9)
Eb (i) = Pb (i) −Pt (15, i) (10)
'Eb = Σ (Eb (i)) ² (i = 0 to 11) (11)
From equation (9) and equation (11), the error function 'Ew for the entire music is expressed by equation (12).
'Ew = 0.7 *' Ea + 0.3 * 'Eb (12)
In the equation (12), when the value of the error function 'Ew approaches as much as 0, the appearance probability of each pitch name for each temporary section of the music to be calculated and the key of each temporary section represented by the temporary music information The appearance probabilities of the pitch names included in the template coincide with each other. That is, when the value of the error function 'Ew approaches as much as possible, the key of each temporary section of the music to be calculated matches the key of each temporary section represented by the temporary music information. Thus, the error function is an index indicating the degree of difference between the key of each temporary section of the music to be calculated and the key of each temporary section represented by the temporary music information.

  Following step S403, the music information calculation unit 15 determines whether or not the value of the error function calculated in step S403 is smaller than a predetermined value (step S404). The predetermined value is set to a value that is allowed as the key of each temporary section of the music to be calculated matches the key of each temporary section represented by the temporary music information. If the value of the error function is smaller than the predetermined value in step S404, the key indicated by each temporary section of the music to be calculated matches the key of each temporary section represented by the temporary music information. Proceed to step S405. In step S405, the music information calculation unit 15 calculates temporary music information as music information of the music to be calculated. If the value of the error function is larger than the predetermined value in step S404, the process proceeds to step S406. In step S406, the music information calculation unit 15 sets the error function value calculated in step S403 to an initial value 'Ew (0)'.

  After step S406, the music information calculation unit 15 replaces an arbitrary section in the music to be calculated (step S407). That is, the music information calculation unit 15 replaces a part of the music to be calculated in the music. Here, it is assumed that an arbitrary section selected at random is replaced. Specifically, for example, it is conceivable to replace sections in which the start position of the section and the length of the section are selected at random. In the example of FIG. 7, as replacement (1), the D section from 0.0 to 0.2 and the Cm section from 0.7 to 0.9 are switched. The music after replacement (1) is the music shown after replacement (1). After step S407, the music information calculation unit 15 calculates the appearance probability of each pitch name for each temporary section for the music whose section has been replaced (step S408). After step S408, the music information calculation unit 15 sets the appearance probability of each pitch name in each temporary section of the music whose section has been replaced, and each pitch name included in the key template of each temporary section represented by the temporary music information. Using the appearance probability, an error function for the entire music is calculated (step S409). About the calculation method of an error function, it calculates with the method similar to step S403.

  After step S409, the music information calculation unit 15 determines whether or not the value of the error function calculated in step S409 is smaller than the initial value (step S410). After the replacement (1) shown in the example of FIG. 7, in the first half section from 0.0 to 0.7, the ratio of the length of all sections of D to the length of sections of Cm is D: Cm = 6: 1. It has become. In the latter half section from 0.7 to 1.0, the ratio of the length of the Cm section to the length of all the sections of D is Cm: D = 2: 1. That is, the music after the replacement (1) is closer to the music of the temporary music information ([0.0-0.7: 2], [0.7-1.0: 15]) than before the replacement. Yes. For this reason, it is understood that the value of the error function calculated in step S409 is smaller than the initial value 'Ew (0) set in step S406. That is, if the error function after the replacement (1) is ‘Ew (1)’, ‘Ew (1) <’ Ew (0). In this case, since the value of the error function is smaller than the initial value, the process proceeds to step S412. When the value of the error function is greater than the initial value 'Ew (0) or more, the music information calculation unit 15 restores the section replaced in step S407 (step S411), and again selects another section in step S407. Replace.

  Following step S410, the value of the error function calculated in step S409 and the information of the replaced section are stored in the replacement information storage unit 16 (step S412). Here, FIG. 8 shows an example of information stored in the replacement information storage unit 16. As shown in FIG. 8, the replacement information storage unit 16 stores, as replacement information, a replacement number n (n is a natural number) indicating the replacement order, information on the replaced section, and the value of the error function after replacement. ing. In the replacement (1) in FIG. 7, the section D from 0.0 to 0.2 and the section Cm from 0.7 to 0.9 are replaced. Therefore, in step S412, the replacement information storage unit 16 stores information on the sections [0.0−0.2] and [0.7−0.9] with respect to the replacement number (1) and the error function 'Ew. (1) is stored.

  After step S412, the music information calculation unit 15 determines whether or not the stored error function value is smaller than a predetermined value (step S413). This predetermined value is the same value as the predetermined value in step S404. If the value of the error function is larger than the predetermined value in step S413, the process returns to step S407, and the section is further switched. Here, it is assumed that the value of the error function after the replacement 1 shown in FIG. 7 is greater than a predetermined value and the process returns to step S407. Then, as shown in FIG. 7, in step S407, as the second section replacement (replacement (2)), the section of Cm from 0.5 to 0.6 and the section of D from 0.9 to 1.0 Suppose you want to swap sections. At this time, as shown in FIG. 7, it can be seen that the music after the replacement (2) indicating the replacement (2) is the same as the music in the provisional music information. Accordingly, in this case, in step S410, it is determined that the value is smaller than the previous time (error function value 'Ew (1) after replacement 1), and replacement information related to replacement (2) is stored in replacement information storage unit 16 in step S412. The

  Further, the error function value 'Ew (2) after replacement (2) is 0 and smaller than a predetermined value, and the key indicated by each section of the music whose section has been replaced and the key of each temporary section represented by the provisional music information Since they match, the process proceeds to step S414. At this time, the music information of the music after the replacement (2) matches the provisional music information. In step S414, the music information calculation unit 15 calculates the music information of the music to be calculated based on the replacement information stored in step S412. Specifically, as shown in FIG. 9, the replacement numbers stored in the replacement information storage unit 16 are traced in the reverse order, and the temporary music information, that is, the music information after the replacement (2) is replaced. FIG. 9 is a diagram schematically showing the process of step S414. In FIG. 8, the replacement information storage unit 16 stores replacement numbers (1) and (2) and information of replacement sections corresponding to the numbers. Therefore, in step S414, as shown in FIG. 9, first, for the music information after the replacement (2), based on the information of the replacement section of the replacement number (2), the section [0.5-0.6] and the section [ 0.9-1.0] (after replacement (2) in FIG. 9). Thereafter, based on the information of the replacement section with the replacement number 1, the section [0.0-0.2] and the section [0.7-0.9] are replaced (after replacement (1) in FIG. 9). As shown in FIG. 9, it can be seen that the music information after the replacement 1 matches the music information of the actual music to be calculated. That is, the music information calculation unit 15 calculates the music information after the replacement (1) as the music information of the music to be calculated.

  As described above, the music information calculation unit 15 repeatedly performs the process of exchanging the sections of the music to be calculated, thereby reducing the error function. That is, the music information calculation unit 15 performs a process of bringing the music to be calculated closer to the music of the temporary music information. In this process, the music information calculation unit 15 records the two replaced sections in order in the replacement information storage unit 16. When the error function becomes smaller than the predetermined value, the music information calculation unit 15 determines that the music information of the music to be calculated that has been processed is provisional music information. Thereafter, the music information calculation unit 15 replaces the sections that have been replaced so far with respect to the confirmed music information in the reverse order. When the key is replaced until the end, the key of each section represented by the confirmed music information matches the key of each section of the actual music to be calculated, and the music information calculation unit 15 sets the music information of the actual music to be calculated. Can be calculated.

  In steps S407 to S413 described above, an arbitrary section selected at random as the section replacement method is replaced. However, the present invention is not limited to this. For example, the music information calculation unit 15 uses, for example, “section 1” and “section 2” as the two sections to be exchanged by using a genetic algorithm or the like with the error function 'Ew as the fitness, and [start position of section 1] , Start position of section 2, length of section]. That is, the music information calculation unit 15 expresses that N (natural number) replacement is performed continuously by creating N sets of [start position of section 1, start position of section 2, length of section]. , Each parameter is optimized so that the error function 'Ew (N) resulting from the N-time replacement is smaller than a predetermined value, and the result is used to replace the interval. The music information calculation unit 15 may employ such a replacement method.

  Further, for example, for the music to be calculated to be processed, among the temporary sections represented by the temporary music information assumed in step S401, the sections corresponding to the sections extracted from the temporary sections of different keys are replaced. May be. In the example of FIG. 7, with respect to the music to be processed, sections corresponding to some sections within the temporary section [0.0-0.7] of D in the music of the temporary music information; Sections corresponding to some sections in the temporary section [0.7-1.0] of Cm up to 1.0 are replaced. This is because the error function is calculated by comparing the music to be processed and the music of the temporary music information in each temporary section. For example, for music to be processed, sections (for example, [0.0-0.2] and [0.5-0] corresponding to some sections in the temporary section [0.0-0.7] of D). .7]) is replaced. Here, if there are many D keys in the [0.0-0.7] temporary section of the music to be processed, the value of the error function is small. However, even if a part of the [0.0-0.7] temporary section of the music to be processed is replaced, the D key included in [0.0-0.7] of the music to be processed is replaced. The proportion of s does not change. That is, even if a part of the [0.0-0.7] temporary section of the music to be processed is replaced, the value of the error function does not change. Therefore, for the music to be calculated to be processed, among the temporary sections represented by the temporary music information assumed in step S401, the sections corresponding to some sections respectively extracted from the temporary sections of different keys are replaced. By doing so, the value of the error function can be efficiently reduced.

  In the above description, the music information is represented by [section start position-section end position: key j of the section]. However, when music information is displayed to the user, as shown in FIG. The labels may be displayed using different labels for each key. FIG. 10 is a diagram illustrating an example of a label in which music information is displayed in a different pattern for each key. Thereby, the user can easily determine the structure of the music and the visibility is improved.

  In the above description, the music information related to the key is calculated. However, the music information related to the tonality and the scale may be further calculated using the music information related to the key. For example, in the case of calculating music information related to tonality, as shown in FIG. 4 described above, 24 kinds of tones are major tones (j = 0 to 11) or minor tones (j = 12 to 23). Since it is classified into two types, the key included in the music information may be classified as major or minor based on the type. Thereby, the user can grasp a section in which a major key or a minor key exists. In addition, when calculating music information related to a scale, the key of each section may be classified into one of the scales. Here, the relationship between the 24 keys and the scale is, for example, the relationship shown in FIG. FIG. 11 is a diagram illustrating an example of the relationship between 24 types of keys and scales. In FIG. 11, a scale number s (s = 0 to 11) is assigned to each scale. As shown in FIG. 11, it can be seen that two of the 24 tones have the same scale, and there are a total of 12 scales. For example, the two types of keys C (j = 0) and Am (j = 12) have the same scale, and the scale number s is s = 0. Thus, by classifying the key of each section into one of the scales (s = 0 to 11) based on the relationship between the scale and key shown in FIG. 11, the user can adjust the major key and the minor key. Can be grasped.

  As described above, the music information calculation apparatus 1 according to the present embodiment does not collate the music appearance probability distribution with the template appearance probability distribution on a one-to-one basis. The ratio (degree of load) occupied by is calculated. As a result, it is possible to accurately detect the key of the music even when the music for detecting a plurality of keys included in the music is composed of a plurality of types of keys including modulation. And the music information calculation apparatus 1 which concerns on this embodiment calculates the music information showing each key which comprises the music detected correctly, and the section in which each said key exists, and this music information is correctly calculated. Of course, it is possible to present a section in which a plurality of tones and other types (for example, tonality) included in the music exist, to the user. Thereby, the user can grasp | ascertain correctly the area where several tones and other types (for example, tonality etc.) contained in a music exist.

  In the above description, the template storage unit 13 stores only one template for each of the 24 types of keys. However, a plurality of templates may be stored for one type of key. For one type of key, for example, a template is prepared for each genre such as pop, jazz, and classic, and is stored in the template storage unit 13. And the music information calculation apparatus 1 can calculate correctly the music information according to the genre of music by calculating the load degree about all the templates memorize | stored in the template memory | storage part 13. FIG. Further, since the load level of each template includes genre information, it is possible to calculate the genre of music.

  Further, for one type of key, a template may be prepared for each specific part, such as a melody and a bass, and stored in the template storage unit 13. And the music information calculation apparatus 1 calculates the load degree about the template matched with the part of the music input among the templates memorize | stored in the template memory | storage part 13, The music according to the part of the music input Information can be calculated accurately.

  Further, as a template, an appearance probability distribution corresponding to tonality, scale, or chord may be used instead of 24 types of tones. Then, by calculating the degree of load for each template, music information regarding tonality, scale, or chord can be directly calculated.

  In addition, the scale information that receives input by the scale information input unit 11 described above may include at least one performance part such as a melody or bass, and may be scale information including a plurality of parts, for example. Good. Further, the performance time of the music constituting the scale information may be the entire performance time or a part of the performance time. Here, the part of the performance time may be, for example, the performance time of the first half of the music excluding the prelude. This is because the performance time is generally composed of major keys. As a result, the music information calculation apparatus 1 can calculate music information with higher accuracy. Moreover, the processing burden of the music information calculation apparatus 1 can be reduced.

  As shown in FIG. 12, the music tone information detection unit 17 according to the present embodiment is further provided, and the music tone of the music is calculated using the load degree calculated by the load degree calculation unit 14 and the music information calculated by the music information calculation unit 15. Information may be detected. FIG. 12 is a block diagram showing a configuration in which the tune information detecting unit 17 is further provided in the tune information calculating apparatus 1. Here, the “musical tone information” is information indicating an index indicating the musical tone of the entire music to be detected, not each section. The tone information includes information indicating 24 load degrees as a set (a set of load degrees) or various information calculated from the set of load degrees. Here, the various information is information indicating, for example, tone, tonality occupancy, tone, scale, and the like. The “tonal occupancy rate” means the occupancy rate of major (or minor) in the music.

  First, a method for detecting music tone information using the load degree calculated by the load degree calculation unit 14 by the music tone information detection unit 17 will be described. The tune information detection unit 17 detects the tune information of the music based on the 24 load degrees calculated by the load degree calculation unit 14. Specifically, the music tone information detection unit 17 includes information (a set of load degrees) indicating 24 load degrees as a set, a tone calculated from the set of load degrees, a tonality occupancy rate, a tonality, and A scale or the like is detected as tune information. In the following, the tone, tonality occupancy, tonality, and scale detection methods will be described.

First, for the key, the music tone information detection unit 17 obtains the key of the key number j corresponding to the template having the maximum load degree W (j), and detects the key as the most dominant key. Next, for the tonal occupancy rate, the tune information detecting unit 17 detects the major occupancy rate and the minor occupancy rate in the music as the tonality occupancy rate by the following method. If the major occupancy rate is Rmaj and the minor occupancy rate is Rmin, the key numbers j = 0 to 11 in FIG. 4 are major and j = 12 to 23 are minor, so Rmaj and Rmin are as follows. It is calculated by the following formula.
Rmaj = ΣW (j) (j = 0 to 11) (13)
Rmin = ΣW (j) (j = 12 to 23) (14)
Therefore, the music tone information detection unit 17 detects Rmaj and Rmin calculated by the equations (13) and (14) as tonality occupancy rates.

  Next, for the tonality, the tune information detecting unit 17 detects the dominance of the major or minor tonality as the tune information. That is, the music tone information detection unit 17 compares the value of Rmaj calculated by the above formulas (13) and (14) with the value of Rmin, and if the value of Rmaj is large, the music tone information detection unit 17 If is large, minor tonality is detected as music tone information.

Next, a scale detection method will be described with reference to FIG. In FIG. 11, two of the 24 tones have the same scale, so there are a total of 12 scales. Each scale is given a scale number s (s = 0 to 11) as shown in FIG. Here, assuming that the occupancy ratio of each scale in the music is the scale occupancy ratio Rs (s) (s = 0 to 11), Rs (s) is calculated as follows.
Rs (0) = W (0) + W (12)
Rs (1) = W (1) + W (13)
Rs (2) = W (2) + W (14)
・
・
・
Rs (11) = W (11) + W (23)
Therefore, the music tone information detection unit 17 obtains the scale of the scale number s that maximizes the value of the scale occupation ratio Rs (s), and detects the scale as the most dominant scale.

  Next, a method for detecting music tone information using the music information calculated by the music information calculating unit 15 by the music information detecting unit 17 will be described. The music information calculated by the music information calculation unit 15 is information representing each key constituting the music and a section in which each key exists, as shown in the bottom of FIG. The music tone information detection unit 17 sums up the lengths of the sections showing the same key from the music information calculated by the music information calculation unit 15 and obtains the key with the maximum length of the section. Detect as a key. The music tone information detection unit 17 uses the same tonality based on the relationship between the tone and the tone shown in FIG. 4 described above from the tone of each section represented in the song information calculated by the song information calculation unit 15. The tones having the longer section length are detected as music tone information. The music tone information detection unit 17 uses the same tonality based on the relationship between the tone and the tone shown in FIG. 4 described above from the tone of each section represented in the song information calculated by the song information calculation unit 15. The lengths of the sections indicating the tones are totaled, the ratio of the major and minor tones in the music is calculated, and the tonality occupancy is detected. Moreover, the music tone information detection part 17 shows the same scale based on the relationship between the key and scale shown in FIG. 11 mentioned above from the key of each area represented by the music information calculated by the music information calculation part 15. The lengths of the sections are obtained by summing up the lengths of the sections, and the scale is detected as the most dominant scale.

  As shown in FIG. 12, by further providing the music tone information detection unit 17, the user can grasp the image of the entire music without actually listening to the music. Specifically, the user can grasp the ratio of a plurality of keys included in the entire music piece by the set of load degrees detected by the music tone information detection unit 17. That is, when the value of the load degree of one key is large and the value of the load degree of the other keys is small, the user can grasp that the entire music is a monotone music. Further, when the value of the load degree is approximated by a large number of keys, the user can grasp that the entire music is a music composed of a large number of keys. In addition, the user can grasp the image of the entire music without actually listening to the music based on the music tone information such as the detected key, tonality occupancy, tonality, and scale.

  Note that the music information calculation apparatus 1 according to the present embodiment may be realized by causing a general computer system 100 to execute a music information calculation program. FIG. 13 is a block diagram illustrating a configuration example in which the music information calculation apparatus 1 is realized by the computer system 100. In FIG. 13, the scale information input unit 11, the appearance probability calculation unit 12, the template storage unit 13, the load degree calculation unit 14, the music information calculation unit 15, and the replacement information storage unit 16 are configured as shown in FIG. 1. The same reference numerals are assigned to the parts and the functions are the same, and the description is omitted.

  In FIG. 13, the computer system 100 includes an arithmetic processing unit 110, a storage unit 120, and a disk drive device 130. The arithmetic processing unit 110 is configured by a CPU or a memory, and is the same as the scale information input unit 11, the appearance probability calculation unit 12, the load degree calculation unit 14, and the music information calculation unit 15 by executing a music information calculation program. Realize the function. The storage unit 120 is configured by a recording medium such as a hard disk, and realizes the same functions as the template storage unit 13 and the replacement information storage unit 16 by executing a music information calculation program. The disk drive device 130 reads the music information calculation program from the recording medium 140 in which the music information calculation program for causing the computer system 100 to function as the music information calculation device 1 is stored. When the music information calculation program is installed in any computer system 100, the computer system 100 can function as the music information calculation device 1 described above. The recording medium 140 is a recording medium in a format that can be read by the disk drive device 130 such as a flexible disk or an optical disk. Moreover, the music information calculation program may be installed in the computer system 100 in advance.

  In the above description, the music information calculation program is provided by the recording medium 140, but may be provided by an electric communication line such as the Internet. Further, the processing in the music information calculation may be a form in which all or a part is processed by hardware.

(Second Embodiment)
Next, a music search apparatus 2 according to the second embodiment of the present invention will be described with reference to FIGS. 14 and 15. FIG. 14 is a block diagram showing a music search device 2 according to the second embodiment of the present invention. 14, the music search device 2 includes a music input unit 21, a scale data conversion unit 22, an appearance probability calculation unit 12, a music data storage unit 23, a template storage unit 13, a load degree calculation unit 14, a music information calculation unit 15, A replacement information storage unit 16, a music tone information detection unit 17, a music selection rule storage unit 24, a music selection information input unit 25, and a search unit 26 are provided. In FIG. 14, the appearance probability calculation unit 12, the template storage unit 13, the load degree calculation unit 14, the music information calculation unit 15, the replacement information storage unit 16, and the music tone information detection unit 17 are described in the first embodiment. The same reference numerals are assigned to the components of the music information calculation apparatus 1 and the functions are the same, and the description thereof is omitted.

  FIG. 15 is a diagram illustrating a flowchart showing the flow of processing of the music search device 2 according to the second embodiment. In FIG. 15, steps S <b> 1 to S <b> 4 are denoted by the same reference numerals as steps S <b> 1 to S <b> 4 of the process (see FIG. 5) of the music information calculation apparatus 1 described in the first embodiment, and are the same. The processing is omitted. Hereinafter, the processing flow of the music search apparatus 2 will be described with reference to FIG.

  The music input unit 21 determines whether or not input of music data has been accepted from outside or inside the music search device 2 (step S11). As a result of the determination in step S11, if it is determined that the music data is not input, the process in step S16 is performed. On the other hand, if it is determined as a result of determination in step S11 that music data has been input, processing in step S12 is performed. That is, the music input unit 21 stores the input music data in the music data storage unit 23 (step S12).

  Here, the music data may be audio data or scale data. The scale data is, for example, audio data in PCM format, audio data in compression format such as MP3 and AAC, and the like. The scale data is scale data indicating scale information such as SMF (standard MIDI file). Note that the input music data only needs to include at least one performance part such as a melody or bass, and may be data composed of a plurality of parts, for example. The performance time of the music data may be the entire performance time or a part of the performance time.

  After step S12, when the music data stored in step S12 is voice data (for example, PCM format voice data), the scale data conversion unit 22 converts the voice data into scale data indicating scale information. (Step S13). The scale data conversion unit 22 converts voice data into scale data by a method disclosed in, for example, Japanese Patent Laid-Open No. 58-181090. In addition, when the audio data is compressed audio data such as MP3 and AAC, the scale data converter 22 converts the audio data into PCM audio data, and then converts it into musical scale data. Note that the method of converting to scale data is not limited to the above method, and other methods may be used. When the music data stored in step S12 is scale data such as SMF, the processes in steps S1 to S4 are performed without performing the process in step S13.

  Next to step S13, in steps S1 to S4 (see FIG. 5), music information is calculated based on the scale data stored in step S12 or the scale data converted in step S13. After steps S1 to S4, the music tone information detection unit 17 detects the music tone information of the entire music by the same method as in the first embodiment described above (step S14). In step S15, the music data storage unit 23 stores the calculated music information and the music tone information of the entire music. Here, the song data storage unit 23 stores the song data stored in step S12, the song information of the song data calculated in step S4 shown in FIG. 5, and the tone information of the entire song detected in step S14. Each is associated and stored. FIG. 16 is a diagram illustrating an example of data stored in the music data storage unit 23. In FIG. 16, in the music data storage unit 23, in addition to the music data, the music information, the key (K), the tonality (T), the most dominant scale (S), and the major scale as the music tone information of the entire music are recorded. The occupation rate (Rmaj) is stored. In FIG. 16, the scale (S) of the music data is indicated by a scale number s. The music data stored in the music data storage unit 23 is managed by the music number, and the music data, the music information associated with the music data, and the music tone information of the entire music can be deleted or added as appropriate. is there.

  After step S15, the search unit 26 determines whether or not music selection information is input from the music selection information input unit 25 (step S16). The user inputs music selection information of a desired music using the music selection information input unit 25. FIG. 17 is a diagram illustrating an example of an input screen of the music search device 2. In FIG. 17, a music selection information list 251, a designated section input unit 252, and a search button 253 are displayed on the input screen. Then, the user uses the music selection information input unit 25 to select desired music selection information from the music selection information in the music selection information list 251. In addition, the user uses the music selection information input unit 25 to specify a section of a song to be selected by the specified section input unit 252. Thereafter, the user presses the search button 253 to determine music selection information.

  As a result of the determination in step S16, if the music selection information is not input, the process returns to step S11. On the other hand, if the music selection information is input as a result of the determination in step S16, the process in step S17 is performed.

  After step S16, the search unit 26 specifies a search condition corresponding to the input music selection information (step S17). Here, as a method of specifying the search condition corresponding to the input music selection information, there is a method of specifying based on the music selection rule stored in the music selection rule storage unit 24. FIG. 18 is a diagram illustrating an example of a music selection rule. The music selection rule storage unit 24 stores music selection rules for searching for music. In FIG. 18, the music selection information displayed in the music selection information list 251 and the search conditions corresponding to the music selection information are set in advance as the music selection rules. That is, in FIG. 18, for example, for music selection information “bright”, the search condition is music data whose tonality as music tone information is major, and for music selection information “somewhat fun”, the search condition is major as music tone information. It is set as music data having an occupation ratio Rmaj of 0.6 to 0.8. That is, in the example of the search condition shown in FIG. 18, a condition based on the music tone information is adopted as the search condition.

  The music selection information stored in the music selection rule storage unit 24 may be, for example, the degree of “fun sadness” without classifying “fun”, “slightly fun”, “normal”, “slightly sad”, and “sad” into five levels. Good. At this time, the music selection information is set to a degree sx of “fun (1.0) sad” (0.0), for example. The search condition is set, for example, as music data having a major occupation ratio Rmaj within a degree sx and an error within 0.1. Then, the user inputs music selection information using a bar such as a slider as the music selection information input unit 25.

  Following step S17, the search unit 26 searches the music data stored in the music data storage unit 23 in step S12 based on the search condition specified in step S17, and matches the condition. The name of the song to be displayed is displayed (step S18). Specifically, the search unit 26 first refers to the song information of each song stored in the song data storage unit 23, and is specified in step S17 for the section corresponding to the specified section input in step S16. The music tone information indicated by the search condition is detected. The method for detecting the tone information from the song information in the designated section is the same as the method for detecting the tone information using the song information calculated by the song information calculating unit 15 by the tone information detecting unit 17 in the first embodiment. Is the method. That is, for example, in step S16, the user selects “bright” and designates a section of [0.0−0.3]. At this time, the search condition is tonality [T == Maj] as music tone information as shown in FIG. Therefore, the search unit 26 refers to the music information of each music stored in the music data storage unit 23, and adjusts each section represented by each music information in the specified section [0.0-0.3]. Thus, based on the relationship between the tone and tonality shown in FIG. 4 described above, the lengths of the sections showing the same tonality are summed up to detect the tonality with the longer section length. In this way, the search unit 26 searches for music information that matches the search condition within the specified section. In other words, in the above example, music information in which the tonality section that is major [Maj] in the specified section [0.0-0.3] is longer than the tonality section that is minor [Min] is retrieved. It will meet the conditions. Here, the music data storage unit 23 stores music information and music data in association with each other. Therefore, when the search unit 26 searches for music information that matches the search condition, the music that matches the search condition is searched.

Further, in step S18, a process of reproducing the song title of the displayed song may be further performed. In step S18, only the section corresponding to the specified section that matches the search condition may be reproduced.

  In step S16 described above, the user selects the music selection information in the music selection information list 251 and designates the section of the music to be selected by the designated section input unit 252. For example, only the music selection information in the music selection information list 251 is selected. You may search for music by selecting. In this case, the search unit 26 searches for music that matches the specified search condition based on the tune information of the entire music stored in the music data storage unit 23. At this time, the music information of each music stored in the music data storage unit 23 is not used for the search of the search unit 26, but the music information can be effectively used even in such a case. In other words, using the music information of each music, the section that matches the search condition among the music searched in step S18 is distinguished and displayed. For example, when the music selection information “bright” is selected, the search condition is [T == Maj]. At this time, as shown by the music number 1 in FIG. 16, a music with Rmaj = 0.93, that is, a music with a dark feeling in 7% of the music may be searched. If the first 7% of the song is minor [Min] and the latter 93% is major [Maj], the darker part of the minor [Min] will be reproduced when the song is played from the beginning. As a result, the user may hear the music selection information “bright” so as not to match. Therefore, the music information is displayed to the user so that a section that matches the music selection condition [T == Maj] can be distinguished. As a result, the user can grasp the “bright” section in the music, and the user can be prevented from feeling uncomfortable that he / she hears the song selection information “bright”.

  Moreover, in the above description, the user specified the search condition used for a search by inputting music selection information. Here, in another embodiment, the user may specify the search condition by directly inputting it. For example, in the music selection information input unit 25, the user inputs a condition of “key C” or “major”, a specified section, or a condition that combines a plurality of these conditions. At this time, the search unit 26 searches for music that matches the search condition input by the user using the music information stored in the music data storage unit 23 and the music tone information of the entire music, and the music that matches the search condition. The section that matches the song name and search condition is displayed. As a result, the user can search for music by freely inputting search conditions instead of search conditions based on preset music selection rules.

  After step S18, the user selects whether or not to end the process of the music search device 2 (step S19). If not, the process returns to step S11. The above processing is performed for each input music data.

  As described above, with the music search device 2 according to the present embodiment, the user searches for music based on the music information of the music or the music tone information of the entire music, and the search condition in the music that matches the search condition or in the music It is possible to find a section that matches

  In the above-described step S15, the music data storage unit 23 stores the music information and the music tone information of the entire music such as the set of load degree, key, tonality occupancy, tonality and scale. , Not all information needs to be stored. Only necessary information may be stored in accordance with music selection information and search conditions.

  Note that the music processing devices (the music information calculation device 1 and the music search device 2) described in the first and second embodiments may be configured by an integrated circuit. For example, in the first embodiment, the appearance probability calculation unit 12, the load degree calculation unit 14, and the music information calculation unit 15 may be configured by an integrated circuit. At this time, the integrated circuit includes a music piece, a template stored in the template storage unit 13, an input terminal for inputting replacement information stored in the replacement information storage unit 16, music information calculated by the music information calculation unit 15, And an output terminal that outputs section replacement information that occurs in the calculation process of the music information calculation unit 15. In the second embodiment, the appearance probability calculation unit 12, the load degree calculation unit 14, the music information calculation unit 15, the music tone information detection unit 17, the scale data conversion unit 22, and the search unit 26 are configured by an integrated circuit. May be. At this time, the integrated circuit receives the music, the template stored in the template storage unit 13, the music selection rule, the music selection information, the music data stored in the music data storage unit 23, the music information stored in the music data storage unit 23, and the music tone. An input terminal for inputting information and replacement information stored in the replacement information storage unit 16 is provided. Further, the integrated circuit uses the music information calculated by the music information calculation unit 15, the replacement information of the sections generated in the calculation process of the music information calculation unit 15, the music tone information detected by the music tone information detection unit 17, and the search unit 26. And an output terminal for outputting the retrieved result. In addition, about the component parts (for example, the template memory | storage part 13 and the replacement information memory | storage part 16 etc.) which memorize | store data etc., if it is an integrated circuit provided with a memory | storage part, it memorize | stores in the said memory | storage part suitably, You may make it comprise.

  The music information calculation apparatus according to the present invention can also be applied to uses such as a music search apparatus, a jukebox, and an audio player that search for music using the calculated music information.

The block diagram which shows the music information calculation apparatus 1 which concerns on the 1st Embodiment of this invention. The figure which shows an example of the scale information of the music input into the scale information input part 11 The figure which shows the distribution of the appearance probability concerning each pitch name conceptually The figure which shows an example of the calculation result of the load degree about 24 types of templates and the said each template The figure which shows the flowchart showing the flow of a process of the music information calculation apparatus 1 which concerns on 1st Embodiment. The flowchart which shows the flow of the process which calculates music information The figure which showed typically the data of each music in a part of processing which calculates music information The figure which shows an example of the information memorize | stored in the replacement information storage part 16 The figure which showed the process of step S514 typically. The figure which shows the example of the label in which music information was displayed with the pattern which changes for every key The figure which shows an example of the relationship between 24 types of keys and scales The block diagram which shows the structure which further provided the music tone information detection part 17 in the music information calculation apparatus 1. FIG. 2 is a block diagram showing a configuration example in which the music information calculation apparatus 1 is realized by the computer system 100 The block diagram which shows the music search apparatus 2 which concerns on the 2nd Embodiment of this invention. The figure which shows the flowchart showing the flow of a process of the music search apparatus 2 which concerns on 2nd Embodiment. The figure which shows an example of the data memorize | stored in the music data storage part 23 The figure which shows an example of the input screen of the music selection information input part 25 Diagram showing an example of music selection rules

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Music information calculation apparatus 11 Scale information input part 12 Appearance probability calculation part 13 Template storage part 14 Load degree calculation part 15 Music information calculation part 16 Replacement information storage part 17 Musical tone information detection part 100 Computer system 110 Arithmetic processing part 120 Storage part 130 Disc drive device 140 Recording medium 2 Music search device 21 Music input unit 22 Musical scale data conversion unit 23 Music data storage unit 24 Music selection rule storage unit 25 Music selection information input unit 26 Search unit

Claims (11)

A music processing device that calculates music information representing each type constituting the music part and a section in which each type exists for a music part that is at least a part of the predetermined music,
For the music part, an appearance probability acquisition unit that acquires a distribution of appearance probabilities for each pitch name;
A template storage unit that stores templates for each of a plurality of types, each of which is a template that is different for each type of music and that indicates a distribution of appearance probabilities for each pitch name;
For each of the plurality of templates stored in the template storage unit, the load degree indicating the ratio of the template to the distribution of the appearance probability of each pitch name acquired by the appearance probability acquisition unit, and the load for each type A load degree calculation unit for calculating the degree;
The music information is calculated so that the ratio of the total length of the sections in which the types constituting the music part exist to the whole music part and the magnitude of the load degree of the type are the same for each type. A music processing apparatus comprising: a music information calculation unit for performing The music information calculation unit
For each type constituting the music part, a setting unit that sets provisional music information in which the length of the section in which each type exists is set to a length that matches the magnitude of the load degree of each type,
A replacement unit for replacing a part of the music part in the music part so that the type constituting the music part and the type represented by the temporary music information match in each section represented by the temporary music information;
A replacement information storage unit for storing replacement information indicating the contents replaced by the replacement unit;
The music composition according to claim 1, further comprising: a reverse replacement section that replaces a part of the provisional music information in an order reverse to the order replaced by the replacement section based on the replacement information stored in the replacement information storage section. Processing equipment. The replacement unit calculates the degree of difference between the type represented by the provisional music information and the type constituting the music part in each section represented by the provisional music information, and reduces the degree of difference between the music pieces. Repeat the process of replacing a part of the part in the music part, and terminate the process when the difference is less than a predetermined value,
3. The music processing apparatus according to claim 2, wherein the replacement information storage unit stores the replacement information every time the replacement unit performs a process of replacing a part of the music part. A music data storage unit that stores data of a plurality of music pieces for which the appearance probability acquisition unit acquires the distribution of the appearance probabilities;
The music information storage which stores the music information calculated in the music information calculation unit in the music data storage unit in association with the music corresponding to the music information among the music data stored in the music data storage unit And
The music processing apparatus according to claim 1, further comprising: a search unit that searches for music stored in the music data storage unit using the music information. A designation part input unit for accepting an input for a user to designate a part of the music;
5. The search unit according to claim 4, wherein the search unit searches for music information stored in the music data storage unit using a part of music information specified by the specified partial input unit. 6. The music processing apparatus described. A music selection rule storage unit that stores a music selection rule that associates the music selection information to be input by the user and the conditions related to the music information;
When the music selection information is input by the user, the search unit includes the music information associated with the input music selection information among the music information associated with the music whose data is stored in the music data storage unit. The music processing device according to claim 4, wherein music information satisfying the condition is detected, and the music associated with the detected music information is output as a search result. A method used in a music processing apparatus that calculates music information that represents each type constituting the music part and a section in which each type exists for a music part that is at least a part of the predetermined music,
In the music processing apparatus, templates that are different for each type of music and that indicate the distribution of appearance probability for each pitch name are stored in advance for a plurality of types, respectively.
Appearance probability acquisition step for acquiring the distribution of appearance probability for each pitch name for the music part;
For each of the plurality of templates stored in the music processing device, the load degree indicating the ratio of the template in the distribution of the appearance probability of each pitch name acquired in the appearance probability acquisition step, and the load for each type A load degree calculating step for calculating the degree;
The music information is calculated so that the ratio of the total length of the sections in which the types constituting the music part exist to the whole music part and the magnitude of the load degree of the type are the same for each type. A music processing method including a music information calculation step. The music processing apparatus further stores in advance data of a plurality of music for which the distribution of the appearance probability is acquired in the appearance probability acquisition step,
A music information storing step for storing the music information calculated in the music information calculating step in the music processing apparatus in association with the music corresponding to the music information of the music data stored in the music processing apparatus; ,
The music processing method according to claim 7, further comprising: a search step of searching for music whose data is stored in the music processing apparatus using the music information. A program for causing a computer of a music processing apparatus to calculate music information that represents each type constituting the music part and a section in which each type exists for a music part that is at least a part of the predetermined music. There,
In the music processing apparatus, templates that are different for each type of music and that indicate the distribution of appearance probability for each pitch name are stored in advance for a plurality of types, respectively.
Appearance probability acquisition step for acquiring the distribution of appearance probability for each pitch name for the music part;
For each of the plurality of templates stored in the music processing device, the load degree indicating the ratio of the template in the distribution of the appearance probability of each pitch name acquired in the appearance probability acquisition step, and the load for each type A load degree calculating step for calculating the degree;
The music information is calculated so that the ratio of the total length of the sections in which the types constituting the music part exist to the whole music part and the magnitude of the load degree of the type are the same for each type. A program that causes the computer to execute a music information calculation step.   A computer-readable recording medium on which the program according to claim 9 is recorded. An integrated circuit that calculates music information representing each type constituting the music part and a section in which each type exists for a music part that is at least a part of the predetermined music,
For the music part, an appearance probability acquisition unit that acquires a distribution of appearance probabilities for each pitch name;
For a plurality of templates that are different templates for each type of music and indicate the distribution of the appearance probability for each pitch name, the ratio of the template in the distribution of the appearance probability of each pitch name acquired by the appearance probability acquisition unit is A load degree calculation unit for calculating a load degree for each type,
The music information is calculated so that the ratio of the total length of the sections in which the types constituting the music part exist to the whole music part and the magnitude of the load degree of the type are the same for each type. An integrated circuit comprising a music information calculation unit that performs the operation.

Images