Music analysis device, music analysis method, and program

Abstract

Provided is a music analysis device comprising a processor that executes: a process for analyzing audio data of only a partial section of BPM-fixed music and identifying a beat position or a bar position in the section; and a process for expanding the beat position or the bar position as a beat position or a bar position of the entire BPM-fixed music.

Description

Music analysis device, music analysis method, and program 【0001】 The present invention relates to a music analysis device, a music analysis method, and a program. 【0002】 Techniques for detecting beat positions and the like from the audio signal of a music piece are known. For example, Patent Document 1 describes a technique for analyzing an audio signal to detect a beat position and the probability of the presence of each musical instrument sound. Patent Document 2 describes a technique for detecting a temporary beat position and a snare drum sound generation position of music data, and setting a beat position based on the snare drum sound generation position when the snare drum sound generation position is not the temporary beat position. 【0003】 Japanese Patent Application Laid-Open No. 2010-134231 International Publication No. 2019 / 058543 【0004】 The above techniques analyze the entire audio signal of a music piece to identify the beat position and the bar position based on the beat position. For example, by analyzing the audio signal using machine learning or the like, the accuracy of identifying the beat position and the like can be improved. On the other hand, when analyzing the entire audio signal of a music piece, the longer the music piece, the more processing time is required. 【0005】 Therefore, an object of the present invention is to provide a music analysis device, a music analysis method, and a program capable of identifying accurate beat positions and bar positions while reducing the processing time. 【0006】According to one aspect of the present invention, a music analysis device is provided, comprising a processor that performs the following steps: analyzing audio data of only a portion of a BPM-fixed music to identify beat positions or bar positions in the portion; and expanding the beat positions or bar positions as the beat positions or bar positions of the entire BPM-fixed music. According to another aspect of the present invention, a music analysis method is provided, in which the processor performs the steps of: analyzing audio data of only a portion of a BPM-fixed music to identify beat positions or bar positions in the portion; and expanding the beat positions or bar positions as the beat positions or bar positions of the entire BPM-fixed music. According to yet another aspect of the present invention, a program is provided for causing a computer processor to perform the following steps: analyzing audio data of only a portion of a BPM-fixed music to identify beat positions or bar positions in the portion; and expanding the beat positions or bar positions as the beat positions or bar positions of the entire BPM-fixed music. 【0007】 This figure shows an example of the overall configuration of a system according to an embodiment of the present invention. This is a block diagram showing the schematic functional configuration of the music analysis device in the example of Figure 1. This is a flowchart showing an example of processing when only BPM-fixed music is input as music audio data. This is a diagram conceptually showing an example of BPM-fixed music in an embodiment of the present invention. This is a diagram showing an enlarged view of part A shown in Figure 4. This is a diagram showing an enlarged view of part B shown in Figure 4. This is a flowchart showing an example of processing when both BPM-fixed music and non-BPM music are input as music audio data. 【0008】Figure 1 shows an example of the overall configuration of a system according to an embodiment of the present invention. The system 10 according to this embodiment includes a PC (Personal Computer) 100, a DJ controller 200, and a speaker 300. The PC 100 is a device that stores, processes, and plays back audio data, and is not limited to a PC; it may also be a terminal device such as a tablet or smartphone. The PC 100 includes a display 101 that displays information to the user, and an input device such as a touch panel or mouse that acquires user operation input. The DJ controller 200 is connected to the PC 100 via a communication means such as USB (Universal Serial Bus), and acquires user operation input related to music playback via channel faders, crossfaders, performance pads, jog dials, and various knobs and buttons. Audio data is played back using, for example, the speaker 300. 【0009】 In this embodiment, the PC 100 functions as a music analysis device in the system 10 described above. For example, the PC 100 performs processing on the stored audio data in response to user input during playback of the audio data. Alternatively, the PC 100 may perform processing on the audio data before playback and save the processed audio data. In this case, the DJ controller 200 and speakers 300 do not need to be connected to the PC 100 at the time the processing is performed. In this embodiment, the PC 100 functions as a music analysis device, but in other embodiments, DJ equipment such as a mixer or an all-in-one DJ system (digital audio player with communication and mixing functions) may function as a music analysis device. Furthermore, a server connected to the PC and DJ equipment via a network may function as a music analysis device. 【0010】Figure 2 is a block diagram showing the schematic functional configuration of the music analysis device in the example shown in Figure 1. The PC 100, which functions as a music analysis device, is a computer equipped with a processor such as a CPU (Central Processing Unit) or a DSP (Digital Signal Processor). The processor performs beat position / bar position identification processing 120 and beat position / bar position expansion processing 130 by operating according to a program. The processor may also perform BPM fixed music determination processing 150. The program is read from the storage of the PC 100 or from a tangible and non-temporary medium such as a removable recording medium, or downloaded from a server via a network and expanded into the memory of the PC 100. 【0011】 The beat position / bar position identification process 120 is a process that analyzes the music audio data 110 to identify beat positions or bar positions. Only beat positions may be identified, or both beat positions and bar positions may be identified. Beat positions can be identified, for example, by detecting the sound position of the bass drum or positions with high acoustic energy from the waveform of the audio data. Bar positions can be identified, for example, from the pattern of intensity of sound at each beat position. Note that the method for identifying beat positions and bar positions is not limited to the above examples, and various known methods can be used. Machine learning may also be used. 【0012】 In the beat position / bar position identification process 120 of this embodiment, for music audio data 110 which is a BPM-fixed song, the audio data of only a portion of the song is analyzed to identify the beat position or bar position in that portion. A BPM-fixed song is a song in which the BPM is fixed throughout, and is commonly found in songs produced by programming, such as EDM (Electronic Dance Music). If it is known that the BPM is fixed throughout the entire song, the beat position or bar position identified in a portion of the song, specifically in a portion where the sound of the bass drum is clear and the beats and bars are easy to detect, can be extended to the entire song. 【0013】The beat position / bar position expansion process 130 expands the identified beat position or bar position as the overall beat position or bar position of the song when the beat position / bar position identification process 120 analyzes the audio data of only a portion of the BPM-fixed song and identifies the beat position or bar position. More specifically, since the beat position and bar position are identified by timestamps in a portion of the song, the beat position / bar position expansion process 130 sets the beat position and bar position in the preceding and succeeding sections at the same interval as the timestamp interval of the beat position or bar position in that section. The beat position / bar position expansion process 130 outputs the overall beat position / bar position data 140 of the song. 【0014】 In the beat position / bar position identification process 120 and beat position / bar position expansion process 130 described above, only the audio data of a portion of the BPM-fixed song is analyzed, thus reducing processing time compared to, for example, analyzing the entire song. Furthermore, in BPM-fixed songs, even if the beat positions and bar positions in a portion of the song are expanded to the entire song, the accuracy of the beat positions and bar positions does not decrease, and accurate beat positions and bar positions can be identified. In the case of a BPM-fixed song that contains portions where it is difficult to detect beats or bars, expanding the beat positions and bar positions identified in portions where detection is easier may actually improve the accuracy of the beat positions and bar positions in the entire song. 【0015】 Figure 3 is a flowchart showing an example of processing when only BPM-fixed songs are input as music audio data. In this case, the beat position / bar position identification process 120 analyzes the audio data of only a portion of the song for the input music audio data 110 to identify the beat positions / bar positions (step S101), and the beat position / bar position expansion process 130 expands the identified beat positions / bar positions to the entire song (step S102). 【0016】Figure 4 is a conceptual diagram showing an example of a BPM-fixed song in an embodiment of the present invention. In this embodiment, for a BPM-fixed song, the beat position or measure position is identified by analyzing the audio data of only a portion of the song's MP section, SC. ​​The section SC for analysis is preferably a section in which beats and measures can be easily detected, specifically a section of a predetermined length other than the introduction and verse, or a section of a predetermined length starting from the position where the bass drum or drum spurt is first sounded. 【0017】 Figures 5A and 5B are enlarged views of sections A and B shown in Figure 4, respectively. Section A is the introduction to the song. In this section, no drum part, including the bass drum, is played, making it difficult to detect beats and measures. Section B, on the other hand, is the section where the verse following the introduction ends and the chorus begins. The bass drum is first played in the song at position P in this section. Therefore, in the illustrated example, the section SC for analysis is set with position P as the starting point. In the illustrated example, position P, where the bass drum is first played, is after the introduction and verse, but if, for example, the bass drum or drum part is first played in the introduction or verse, the section SC including the introduction and verse may be set. It is also possible to set the section SC from the beginning of the song. On the other hand, if the bass drum or drum part is not played for some time even after the chorus begins, not only the introduction and verse but also part of the chorus may be excluded from the section SC, and the section SC may be set from the position where the bass drum or drum part is first played in the chorus. 【0018】By setting the SC section to a section other than the introduction or verse, or a section starting from the point where the bass drum or drum part is first played, beats and measures can be easily detected within the SC section, allowing for accurate identification of beat and measure positions within a predetermined length of SC section, regardless of the song. The predetermined length of the SC section can be set in time, such as one minute, or in the number of measures, such as two measures. This allows for the identification of beat or measure positions by analyzing an SC section of the same length regardless of the song length, even in songs with a fixed BPM, reducing processing time and making the processing time for each song approximately the same. 【0019】 Referring again to Figure 2, if the music audio data 110 includes both BPM-fixed and non-BPM-fixed music, the PC 100 processor executes the BPM-fixed music determination process 150. Whether or not a music track is a BPM-fixed music track can be determined, for example, by the music metadata associated with the music audio data 110. In this case, the BPM-fixed music determination process 150 refers to the music metadata. Alternatively, the BPM-fixed music determination process 150 may determine whether or not a music track is a BPM-fixed music track by analyzing the entire music audio data 110 and detecting BPM fluctuations. Since analyzing audio data to detect BPM fluctuations is simpler than analyzing audio data to identify beat positions and bar positions, even if the BPM-fixed music determination process 150 is combined with the beat position / bar position identification process 120 and the beat position / bar position expansion process 130, the processing time may be less than, for example, the case where beat positions and bar positions are identified by analyzing the entire music track. 【0020】If the BPM fixed song determination process 150 determines that the song in the song audio data 110 is not a BPM fixed song, the beat position / bar position identification process 120 may analyze the entire song audio data 110 that is not a BPM fixed song to identify the beat position or bar position. Various known methods can be used for analyzing the entire song to identify the beat position and bar position. Machine learning may also be used. For the song audio data 110 that is not a BPM fixed song, the beat position / bar position expansion process 130 is not executed, and the beat position / bar position data 140 is output by the beat position / bar position identification process 120. Alternatively, for the song audio data 110 that is not a BPM fixed song, the beat position / bar position identification process 120 is also not executed, and the beat position and bar position may be identified by a device other than the PC 100, or by another process executed by the PC 100. 【0021】 Figure 6 is a flowchart illustrating an example of processing when both BPM-fixed and non-BPM-fixed songs are input as music audio data. In this case, first, the BPM-fixed song determination process 150 determines whether the input music audio data 110 is a BPM-fixed song (step S103). As already mentioned, the determination may be made by referring to the metadata of the song associated with the music audio data 110, or by analyzing the entire music audio data 110 and detecting BPM fluctuations. If the song is a BPM-fixed song (YES in step S104), the beat position or bar position of the entire song is determined by analyzing the audio data of only a portion of the song and expanding the identified beat position / bar position to the entire song, similar to the example explained above with reference to Figure 3 (steps S101, S102). If the song is not a BPM-fixed song (NO in step S104), the beat position / bar position determination process 120 analyzes the entire audio data of the song to determine the beat position or bar position (step S105). 【0022】According to the embodiments of the present invention described above, by performing a process of analyzing audio data of only a portion of a fixed-BPM song to identify the beat position or bar position in that portion, and a process of expanding the beat position or bar position as the overall beat position or bar position of the fixed-BPM song, it is possible to accurately identify the beat position or bar position while reducing the processing time compared to analyzing the entire fixed-BPM song. By setting the analysis portion of a fixed-BPM song to a predetermined length portion other than the introduction and verse of the fixed-BPM song, or a predetermined length portion starting from the position where the bass drum or drum spurt is first sounded in the fixed-BPM song, the accuracy of identifying the beat position or bar position through analysis of that portion is increased, and for example, accurate beat position or bar position can be identified by analyzing a portion of the same length regardless of the song. This not only reduces processing time, but also makes the processing time for each song approximately the same length. 【0023】 Furthermore, if the audio data of the input song is not limited to songs with a fixed BPM, a process may be executed to determine whether the song is a song with a fixed BPM. This determination process may include, for example, analyzing the entire audio data of the song to detect BPM fluctuations or referencing the song's metadata. If the song is not a song with a fixed BPM, the analysis may not be performed, or a process may be executed to analyze the entire audio data of the song to identify beat positions or bar positions. This allows for the accurate identification of beat positions and bar positions with an optimized processing time depending on the input song. 【0024】 10...System, 100...PC, 101...Display, 110...Music audio data, 120...Beat position / measure position identification processing, 130...Beat position / measure position expansion processing, 140...Beat position / measure position data, 150...BPM fixed song determination processing, 200...DJ controller, 300...Speaker.

Claims

1. A music analysis device comprising a processor that performs the following processes: analyzing audio data of only a portion of a fixed-BPM song to identify the beat position or bar position in said portion; and expanding said beat position or bar position as the beat position or bar position of the entire fixed-BPM song.

2. The music analysis device according to claim 1, wherein the section is a section of a predetermined length other than the introduction and verse of the BPM-fixed music.

3. The music analysis device according to claim 1, wherein the section is a section of predetermined length that starts from the position where the bass drum is first sounded in the BPM-fixed music.

4. The music analysis device according to claim 1, wherein the section is a section of predetermined length that starts from the position where the drum part is first played in the BPM-fixed music.

5. The music analysis apparatus according to claim 1, wherein the processor further performs a process to determine whether or not the music is a BPM fixed music.

6. The music analysis device according to claim 5, wherein the process for determining whether the music is a BPM-fixed music includes a process for analyzing the entire audio data of the music and detecting fluctuations in BPM.

7. The music analysis apparatus according to claim 5, wherein the process for determining whether the music is a BPM fixed music includes a process for referring to the metadata of the music.

8. The music analysis device according to claim 5, wherein the processor further performs a process to analyze the entire audio data of the music and identify beat positions or bar positions when the music is not a BPM fixed music.

9. A music analysis method in which a processor performs the steps of: analyzing audio data of only a portion of a BPM-fixed music to identify the beat position or bar position in the portion; and expanding the beat position or bar position as the beat position or bar position of the entire BPM-fixed music.

10. A program that causes a computer processor to perform the following processes: analyzing audio data of only a portion of a fixed-BPM song to identify the beat position or bar position in the said portion; and expanding the said beat position or bar position as the beat position or bar position of the entire fixed-BPM song.

Images