GB2606522A - A system and method for generating a musical segment - Google Patents

A system and method for generating a musical segment Download PDF

Info

Publication number
GB2606522A
GB2606522A GB2106611.3A GB202106611A GB2606522A GB 2606522 A GB2606522 A GB 2606522A GB 202106611 A GB202106611 A GB 202106611A GB 2606522 A GB2606522 A GB 2606522A
Authority
GB
United Kingdom
Prior art keywords
musical
user
processor
database
elements
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
GB2106611.3A
Inventor
De Villiers Melvyn
Toulson Robert
G Peters Guillem
Turner Alexander
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Phuture Phuture Ltd
Original Assignee
Phuture Phuture Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Phuture Phuture Ltd filed Critical Phuture Phuture Ltd
Priority to GB2106611.3A priority Critical patent/GB2606522A/en
Publication of GB2606522A publication Critical patent/GB2606522A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H1/00Details of electrophonic musical instruments
    • G10H1/0008Associated control or indicating means
    • G10H1/0025Automatic or semi-automatic music composition, e.g. producing random music, applying rules from music theory or modifying a musical piece
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2210/00Aspects or methods of musical processing having intrinsic musical character, i.e. involving musical theory or musical parameters or relying on musical knowledge, as applied in electrophonic musical tools or instruments
    • G10H2210/031Musical analysis, i.e. isolation, extraction or identification of musical elements or musical parameters from a raw acoustic signal or from an encoded audio signal
    • G10H2210/056Musical analysis, i.e. isolation, extraction or identification of musical elements or musical parameters from a raw acoustic signal or from an encoded audio signal for extraction or identification of individual instrumental parts, e.g. melody, chords, bass; Identification or separation of instrumental parts by their characteristic voices or timbres
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2210/00Aspects or methods of musical processing having intrinsic musical character, i.e. involving musical theory or musical parameters or relying on musical knowledge, as applied in electrophonic musical tools or instruments
    • G10H2210/101Music Composition or musical creation; Tools or processes therefor
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2210/00Aspects or methods of musical processing having intrinsic musical character, i.e. involving musical theory or musical parameters or relying on musical knowledge, as applied in electrophonic musical tools or instruments
    • G10H2210/101Music Composition or musical creation; Tools or processes therefor
    • G10H2210/105Composing aid, e.g. for supporting creation, edition or modification of a piece of music
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2210/00Aspects or methods of musical processing having intrinsic musical character, i.e. involving musical theory or musical parameters or relying on musical knowledge, as applied in electrophonic musical tools or instruments
    • G10H2210/101Music Composition or musical creation; Tools or processes therefor
    • G10H2210/111Automatic composing, i.e. using predefined musical rules
    • G10H2210/115Automatic composing, i.e. using predefined musical rules using a random process to generate a musical note, phrase, sequence or structure
    • G10H2210/121Automatic composing, i.e. using predefined musical rules using a random process to generate a musical note, phrase, sequence or structure using a knowledge base
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2210/00Aspects or methods of musical processing having intrinsic musical character, i.e. involving musical theory or musical parameters or relying on musical knowledge, as applied in electrophonic musical tools or instruments
    • G10H2210/155Musical effects
    • G10H2210/245Ensemble, i.e. adding one or more voices, also instrumental voices
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2240/00Data organisation or data communication aspects, specifically adapted for electrophonic musical tools or instruments
    • G10H2240/005Data structures for use in electrophonic musical devices; Data structures including musical parameters derived from musical analysis
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2240/00Data organisation or data communication aspects, specifically adapted for electrophonic musical tools or instruments
    • G10H2240/075Musical metadata derived from musical analysis or for use in electrophonic musical instruments
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2240/00Data organisation or data communication aspects, specifically adapted for electrophonic musical tools or instruments
    • G10H2240/095Identification code, e.g. ISWC for musical works; Identification dataset
    • G10H2240/101User identification
    • G10H2240/105User profile, i.e. data about the user, e.g. for user settings or user preferences
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10HELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2240/00Data organisation or data communication aspects, specifically adapted for electrophonic musical tools or instruments
    • G10H2240/121Musical libraries, i.e. musical databases indexed by musical parameters, wavetables, indexing schemes using musical parameters, musical rule bases or knowledge bases, e.g. for automatic composing methods
    • G10H2240/131Library retrieval, i.e. searching a database or selecting a specific musical piece, segment, pattern, rule or parameter set
    • G10H2240/141Library retrieval matching, i.e. any of the steps of matching an inputted segment or phrase with musical database contents, e.g. query by humming, singing or playing; the steps may include, e.g. musical analysis of the input, musical feature extraction, query formulation, or details of the retrieval process

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Electrophonic Musical Instruments (AREA)

Abstract

A system 10 and method for generating a musical segment, the system comprising a database 14 configured to store a plurality of musical elements in a plurality of categories such as melodic, vocal and rhythmic. A processor 12 is operably connected to the database, ideally via a communications network, and is configured to analyse a primary musical element of one of the plurality of categories and identify at least one musical element from another of the plurality of categories stored in the database that matches one or more predefined parameters of the primary musical element. The identification may be carried out using artificial intelligence. The primary musical element is then combined with the at least one identified element to generate a musical segment, which may involve aligning the tempo and key of the elements. Each musical element in the database may be associated with a user, to whom an alert is sent via the processor when said musical element is used to generate a musical segment. The system may be used to collaboratively compose music.

Description

A SYSTEM AND METHOD FOR GENERATING A MUSICAL SEGMENT
The present invention relates to a system and method for generating a musical segment.
In embodiments, the present invention relates to a system and method for generating a musical segment from separate musical elements.
A "hook" is a musical phrase, often with vocals, that engages the attention of the listener and typically forms a small but important part of a musical number. Typically, a hook comprises a beat (or rhythmic beat such as drums), a melody, and vocals. But in some examples, a hook may comprise only two, or even one, out of the three aforementioned elements.
A hook may be of any length. Typically, in accordance with musical theory, a musical phrase may be four bars long, with each bar comprising four beats (i.e. a four bar phrase in 4/4 time). For example, in the song "Can't Get You Out Of My Head" by Kylie Minogue, the hook is usually identified as the four bars containing the lyrics "La la la, la la la-la la" repeated twice. In other examples, the hook may simply be a part of one or more bars, such as the two beats containing the lyrics "Woo hoo" in the song "Song 2" by Blur. In the case of "Song 2", it is also an example where the hook has no discernible melody, although it does have a discernible pitch variation. In another example, in "We Will Rock You" by Queen, one of the hooks may be identified as the twice-repeated four bar drum and hand clapping introduction.
In the past, music is typically written by one or two individuals, who would create all the individual elements that go into a musical number. For example, in the case of a duo, one person may write the music and beat, and the other may write the lyrics. However, in modern times, as music becomes more and more complex, with an individual musical number being comprised of a large number of components and tracks (e.g. the main melodic track, a bass/beat track, accompanying instrumental tracks, main vocal track, backing vocal track, verse lyrics, chorus lyrics, etc.), writing credits for a musical number often go to several people, in some cases up to and beyond a dozen individuals. For example, the writing credits to "Uptown Funk" by Mark Ronson (ft Bruno Mars) include eleven individuals. It is because of this complexity that the task of finding individuals with the correct talent required to create the different parts of a -2 -successful musical number has also become more and more complex. A talented composer may have a potentially successful melody whilst not able to find the correct people to provide the necessary beat and/or lyrics to create an award winning song.
There is therefore a need for a system and method that would enable composers, lyricists, and songwriters in general to collaborate more easily and in a more efficient manner.
According to an aspect of the present invention, there is provided a system for generating a musical segment, comprising a database configured to store a plurality of musical elements in a plurality of categories, the categories including any of melodic, vocal, and rhythmic; a processor operably connected to the database, configured to analyse a primary musical element of one of the plurality of categories, identify at least one musical element from another of the plurality of categories stored in the database that matches one or more predefined parameters of the primary musical element, and combine the primary musical element with the at least one identified element to generate a musical segment.
The present invention therefore provides a system that combines computer algorithm and human creativity in the generation of sonically pleasing musical segments. The present invention also enables rapid searching of a large database of stored elements, a task that would otherwise take a human an unfeasible amount of time.
In embodiments, the processor and the database are operably connected via a network, and the system is configured to communicate with respective user devices of a plurality of users via the communications network.
In embodiments, the processor is configured to receive a musical element from a user and store the received element in the database under one of the plurality of categories. 3 0
In embodiments, the processor is configured to identify the at least one element from another of the plurality of categories by using artificial intelligence to identify the one or more matching predefined parameters. Preferably, the processor is configured to, when more than one element from the another of the plurality of categories is identified as having one or more matching predefined parameters, assigning a score to each of -3 -the identified elements and using the identified element with the highest score for the generating the musical segment.
In embodiments, the processor is configured to generate the musical segment by aligning the tempo and key of the elements.
In embodiments, each musical element in the database is associated with a user, and the processor is configured to send an alert to each user associated with the elements used to generate the musical segment In embodiments, the primary element is a melodic element, and the at least one predefined parameter includes at least one of a tempo pattern, a pitch pattern, and a musical key pattern.
In embodiments, each user is associated with a user profile stored on the database.
In embodiments, the primary musical element is received from an engaged user. Preferably, each user profile includes a set of user preferences, wherein the processor is configured to identify the at least one element from another of the plurality of categories based on the user preferences of the engaged user.
In embodiments, the processor is configured to receive from the engaged user a user preference score for the generated musical segment, analyse the generated musical segment using artificial intelligence, and update the user preferences based on the analysis and the user preference score.
According to another aspect of the present invention, there is provided a method of generating a musical segment, comprising storing a plurality of musical elements in a plurality of categories in a database, the categories including any of melodic, vocal, and rhythmic; using a processor, analysing a primary musical element of one of the plurality of categories, identifying at least one musical element stored in the database from another of the plurality of categories that matches one or more predefined parameters of the received musical element, and combining the received musical element with the at least one identified element to generate a musical segment. -4 -
Embodiments of the present invention will hereinafter be described by way of examples, with references to the accompanying drawings, in which: Figure 1 is a schematic diagram of a system for generating a musical segment.
The present invention is generally directed to a system for generating a musical segment. A musical segment is a short passage of music that can be self-contained or forms a part of a longer musical number. Generally, a musical segment can be divided into elements including a melody and a beat, and optionally vocals (i.e. lyrics sung by a person). In some examples, a musical segment may comprise a beat and sounds with pitch and/or dynamic variations that are complimentary to the beat. A musical segment may be of any length, from one bar to several bars. For example, a musical segment may be four bars long, with each bar having four beats. In another example, a musical segment may be just one bar long, containing two beats. The system of the present invention enables a user to communicate (e.g. uploading, submitting, or otherwise providing) a musical element of one category, such as a beat, a melody, or vocals, that has the potential to form a part of a musical segment, to the system, and using the capabilities of the system to search for and identify sonically compatible or complimentary elements from other categories on the database that other users have uploaded, and to generate a musical segment from a combination of the user's element and the identified sonically compatible or complimentary element(s). Optionally, the system is able to allow a user to rate the generated musical segment, analyse the generated musical segment along with the user's rating using artificial intelligence, store the results of the analysis as the user's preferences in a user profile in the database, and use the user's preferences in subsequent searches for sonically compatible or complimentary elements. Also optionally, the system is able to alert the users whose uploaded elements have been identified and used in the generation of a musical element, and provide a communication channel to enable the users to connect.
Embodiments of the system of the present invention generally includes a database and a processor that are operably connected together. The database may be any suitable computer implemented database and the processor may be any suitable computer processing unit such as a central processing unit (CPU). In some embodiments, the processor may comprise of multiple components such as a CPU operably connected to a graphics processing unit (GPU) or a field programmable gate array (FPGA). The processor is communicatively connected to the database, either directly or indirectly, -s -such as over a communications network (e.g. over the internet). Users may connect to the system from user devices such computers, mobile phones, or any other suitable communications devices.
In various embodiments, the database is configured to store musical elements that users upload. The uploaded elements are categorised, either based on users' input or the processor is configured to analyse each uploaded element (such as by using artificial intelligence or any other suitable form of machine learning) to determine a suitable category under which to store the uploaded element. Each uploaded element is also stored alongside a record (e.g. in the form of metadata) of the user that uploaded the element. In embodiments, the system is further configured to store user profiles in the database. For example, for each unique user, the system may store one or more of the user's name, a user ID, a user category (e.g. a composer, a musician, a vocalist, a producer, etc.), the category of elements that the user intends to upload (e.g. rhythmic elements, melodic elements, vocal elements), the user's contact details (e.g. email address, telephone number, address, etc.), and the user's personal details (such as age, nationality, gender, etc.) in the user profile. In embodiments, the system is further configured to store a record of all the elements that a user has uploaded in the user profile.
Rhythmic elements are those which are predominantly created by percussive musical instruments, such as drums, marimba, congas or tabla or percussion sound samples which have been created or manipulated in audio software or electronically. Melodic elements are those which are predominantly created by melodic musical instruments with harmonic characteristics, such as guitar, piano, violin, trumpet or melodic sound samples which have been created with digital software or electronic methods, such as synthesisers or from musical sound data waveforms. Vocal elements are those which are predominantly created from the human voice or electronic systems which mimic characteristics of the human voice.
In various embodiments, the processor is configured to analyse an element, hereinafter referred to as the primary element, for certain sonic qualities and to use artificial intelligence to search the database to identify one or more elements from differing categories that are sonically compatible or complimentary with the primary element and suitable to be combined to form a sonically pleasing musical segment. For example, the primary element may be a melodic element, and the processor may analyse the -6 -melodic element to identify its key, time signature, patterns in pitch variation, patterns in dynamic (i.e. volume) variation, and any other suitable qualities. Optionally, based on the analysis, the processor may assign a numerical value along a scale to each of the aforementioned sonic qualities for the melodic element. Then, in a step hereinafter referred to as the matching process, the processor may search the database and identify one or more elements from other categories (e.g. an element from the beat category and an element from the vocals category) that have compatible or complimentary sonic qualities. For example, the processor may analyse the elements stored in the database, assign a numerical value along a scale to each of the aforementioned sonic qualities for each element, and identify the elements with values that match the values for the melodic element. In some preferred implementations of the system, the analysis of the elements in the database may be done prior to the matching process to increase the speed of the matching process. For example, each element may be analysed soon after it is uploaded to the system and optionally have values assigned for each of the aforementioned sonic qualities, and the result of the analyses may be stored in the database alongside the element as metadata. This enables a faster matching process by enabling the processor to search the metadata stored in the database to identify matching elements. Once matching elements have been identified, the processor combines the identified elements (e.g. a matching rhythmic element and a matching vocal element) with the primary element to generate a musical segment, which may be stored in the database.
In some embodiments, after a musical segment has been generated, the processor identifies the users associated with the elements that were selected (for example by reading the metadata associated with each of the elements) and sends an alert or notification to the users to notify them that a musical element that the user uploaded has been used to generate a musical segment. The processor may also send a copy of the generated musical segment to each user. For example, once a musical segment has been generated by combining a primary melodic element with a rhythmic element and a vocal element, the processor reads the metadata of the rhythmic element and the vocal element to identify the users that uploaded these elements. The processor then sends an electronic alert, such as an email, a text message, or an alert on a graphical user interface of the system to the user of the rhythmic element and the user of the vocal element to notify them that their musical elements have been used to generate a musical segment. Optionally, the processor is configured to send a copy of the generated musical segment to the user of the rhythmic element and the user of the -7 -vocal element. In some implementation of the system, instead of sending a copy of the generated musical segment itself, the processor may send a URL link to direct the user of the rhythmic element and the user of the vocal element to download the generated musical segment from the database.
In some embodiments, the system is further configured to connect the users that contributed to a musical segment, to facilitate further collaboration for example. To that end, in some of the embodiments, the system is configured to provide a suitable platform for the users to communicate, for example by providing a social network platform. In some other embodiments, the processor may simply send each contributing user the contact details of the other contributing users. In the above example, the processor may send the user of the melodic element the contact details of the user of the rhythmic element and the contact details of the user of the vocal element, and so on.
Referring to figure 1, an embodiment of the system 10 comprises a processor 12 and a database 14, each communicatively connected to a communications network 16. The communications network 16 enables communication of data between the processor 12 and the database 14. Users connect to, and interact with, the system 10 using user devices 18. In some embodiments, user devices 18 connect to the system 10 via the communications network 16 as shown in figure 1. In some embodiments, user devices 18 connect to the system 10 directly. In some other embodiments, user devices 18 connect to the system 10 directly or via the communications network 16. The processor 12 may be any suitable kind of computer processor, such as a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), or a combination thereof. The database 14 comprises storage memory such as conventional hard drives, solid state drives, or any other suitable storage devices, and may be arranged in any suitable way, such as in a redundant array of inexpensive disks (RAID). A user device 18 may be any suitable communication device, such as a computer or a mobile phone, and a user interacts with the system via any suitable interface, such as via software or an app operating on the user device 18 or via a web interface accessed through the user device 18. Each user is given a user profile, stored in the database 14, containing information such as the user's name, age, gender, address, contact details, and other information to identify the user. In some embodiments, the database 14 may be alternatively or additionally communicatively connected to the processor 12 directly. -8 -
The processor 12 is configured to receive content, including musical elements, that users upload to the system 10 and is configured to store the uploaded content in the database 14. For example, a user accesses the system 10 via an app operating on a user device 18. Through the app on the user device 18, the user is able to upload a musical element to the system 10, which is then received by the processor 12. Before storing the uploaded musical element in the database 14, the processor 12 optionally analyses the uploaded musical element to identify certain sonic qualities, and further optionally assigns numerical values along a scale (i.e. a score) to each identified sonic quality. The results of the analysis and the identity of the user that uploaded the musical element are stored as metadata in the database along with the musical element itself. The user's user profile also includes a record of all the musical elements that the user has uploaded to the system 10. The processor 12 is further configured to receive an input from the user as to a category under which the uploaded musical element is to be stored. For example, if the uploaded musical element is a melody, the user may signal via the user device 18 that the musical element is to be stored under the category of "melodic elements" and the processor 12 stores the uploaded element in the database 14 accordingly. If the uploaded musical element is a beat, the user may signal via the user device 18 that the musical element is to be stored under the category of "rhythmic elements" and the processor 12 stores the uploaded element in the database 14 accordingly. If the uploaded musical element is a vocal track, the user may signal via the user device 18 that the musical element is to be stored under the category of "vocal elements" and the processor 12 stores the uploaded element in the database 14 accordingly. Alternatively, the processor 12 is configured to automatically determine the category under which an uploaded musical element is to be stored. For example, the processor 12 may be programmed with artificial intelligence that is able to analyse the uploaded musical element and determine the suitable category.
For the analysis of an uploaded musical element, the processor 12 is configured to use various computer algorithms, including artificial intelligence, to analyse certain sonic qualifies of the uploaded musical element, including, but not limited to, the key, time signature, patterns in pitch variation, and patterns in dynamic (i.e. volume) variation. In some embodiments, the processor 12 is also configured to analyse other signal features of the uploaded musical element, such as the spectral centroid. The analysis of the uploaded musical element enables to the system 10 to assign a score to each -9 -sonic quality for each musical element stored in the database 14 so that a comparison of different musical elements can be made quickly by comparing the scores.
The processor 12 is further configured to enable a user to find musical elements stored in the database 14 that are sonically compatible or complimentary with the musical element that the user has uploaded to the system 10 and to combine the matching musical elements to generate a musical segment. In particular, once a user has uploaded a musical element, the user can designate that musical element as the primary element, or select an element already stored in the database 14 as the primary element, and instruct the processor 12 to perform a matching process whereby the processor 12 searches the database 14 for musical elements in categories other than the category of the primary element that are sonically compatible or complimentary with the primary element. In embodiments in which scores have been given in the metadata to various sonic qualities of each musical element, the processor 12 scans the metadata stored in the database 14 to identify musical elements in the different categories that match the scores of the primary element. Once matching elements are found, the processor 12 optionally processes the elements to make adjustments so that the pitch and beat of the elements are aligned, and combines the elements to generate a musical segment. For example, if a user uploads a melodic element to the system 10, the melodic element is designated as the primary element and the processor 12 searches the database 14, reads the metadata of all the stored musical elements in the "rhythmic" and "vocal" categories to identify matching elements in those categories, processes the matching elements and the primary element to adjust for any pitch, beat, and/or dynamics misalignment, and combines the aligned elements together to generate a musical segment. The generated musical segment is then sent to the user of the primary element. In some instances, more than one element from one of the other categories may be identified as being suitable for combining with the primary element. For example, the processor 12 may identify two matching elements from the "rhythmic" category and three matching elements from the "vocals" category for a particular primary melodic element. In such a case, the processor 12 is configured to generate musical segments in the same way as described above from all the possible combinations of musical elements, i.e. six different musical segments in total from the primary melodic element with the two identified rhythmic elements and three identified vocal elements. The generated musical elements are then sent to the user of the primary element.
-10 -To identify the users of the musical elements that were used to generate the musical segment, the processor 12 reads the metadata associated with each of the musical elements stored in the database to identify the users that uploaded the respective musical elements and retrieves the users' contact details from the user profiles. In some embodiments, the processor 12 then sends an alert to the users to notify the users that a musical element the user has uploaded has been used in a musical segment. The alert may be in any suitable form, such as a text message, an email, or an alert on a user interface of the system 10. Alternatively or additionally, the processor 12 sends a copy of the generated musical segment to the user in accordance with the user's contact details or sends a URL link to the user to enable the user to download the generated musical segment from the database 14.
Optionally, the system 10 is configured to receive, from the user via the user device 18, a score for the or each of the generated musical segments, where the score reflects the user's personal rating for the generated musical segment. In embodiments, the system 10 is configured to store a user's scoring for each generated musical segment in the database 14, for example as part of the user's user profile, and the processor 12 is programmed with artificial intelligence that learns the correlation between the various sonic qualities of the generated musical segments and the user's corresponding scoring to determine the user's preferences. The processor 12 is further configured to use a user's preferences, and optionally with artificial intelligence, when performing the matching process to generate musical segments that are not only sonically pleasing, but also tailored to the user's particular taste and preferences. With these provisions, the system 10 is able to learn and update its database as more and more musical segments are generated, so that with each generated musical segment, the system 10 is better able to understand a user's taste and preferences, and enables subsequent generated musical segments to be better tailored to the user's taste. The processor 12, or the system 10 itself, is optionally further configured to send the or each generated musical segment to the other users that contributed thereto, and further optionally configured to enable the users to score the or each generated musical segment in the same way as described above.
Optionally, the processor 12 (or the system 10) is further configured to send a means of contacting a user whose musical element has been used in a generated musical segment to the other users whose musical elements have been used in the same generated musical element. The means for contacting a user may be, for example, an email address, a telephone number, or, in embodiments in which the system 10 provides a social network platform to enable users to connect, a user identification. In these embodiments, the system 10 is configured to provide the social network platform for users to access on their respective user devices 18, via the communications network 16. The necessary data and computer program for the social network platform may be stored in the database 14 and the computer program may be executed by the processor 12. With the above-described functionality of generating musical segments from a database of musical elements of different categories from different users and the provision of a social network platform to enable different users to connect, these embodiments enable users that otherwise would not have had the opportunity to meet to collaborate and create music.
Figure 2 shows another embodiment of a system 20 for generating a musical segment. The system 20 includes a processor 22 and a database 24, each communicatively connected to a communications network 26, similar to the system 10 shown in figure 1.
Users connect to, and interact with, the system 20 via the communications network 26 using user devices 28 similar to the system 10 shown in figure 1. The system 20 implements a backend of a broader music platform 30 and the user device 28 implements a frontend. Similar to the system 10 shown in figure 1, the processor 22 of the system 20 shown in figure 2 may be any suitable kind of computer processor, such as a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), or a combination thereof. The processor 22 is also configured to execute artificial intelligence algorithms. The database 24 comprises storage memory such as conventional hard drives, solid state drives, or any other suitable storage devices, and may be arranged in any suitable way, such as in a redundant array of inexpensive disks (RAID). The user device 28 may be any suitable communication device, such as a computer, a tablet, or a mobile phone, and a user interacts with the system 20 via any suitable interface, such as via software or an app operating on the user device 28 or via a web interface accessed through the user device 28. In some embodiments, the database 24 may be alternatively or additionally communicatively connected to the processor 22 directly.
The frontend of the platform 30 is provided by the user device 28 and includes an audio recording feature to allow a user to sing or perform music into a device microphone (not shown) and capture a vocal musical performance to create a musical element which is then stored as an audio file on the user device 28. The user may choose to -12 -perform and record whilst listening to a metronome, pitched drone, or other musical backing track. In some alternative embodiments, the user may access a pre-recorded audio file of a musical element from the memory of the user device 28 or a connected cloud storage location. Recorded or pre-recorded audio files can be reviewed (listened to), edited, and processed in the frontend provided by the user device 28. For example, the audio file may be edited to a chosen length or duration in order to ensure that a suitable musical element is created. For example, a suitable musical element may be one which can be play continuously from start to finish and immediately repeated without hearing glitches or step changes in the timing or volume of the audio.
The frontend provided by the user device 28 also allows users to set the tempo of the musical element and the number of musical bars to be included. For example, a user may select a looping playback on 2, 4 or 8 musical bars. Recorded or pre-recorded audio can also be processed in the frontend to creatively enhance the sonic characteristics of the audio. Audio processing may include, for example, dynamic range compression, addition of reverb and echo, spectral processing (for example with audio equalisation tools), and established artistic audio processing tools such as autotune, doubling, phasing, flanging or distortion. Additional automated audio processing may be implemented, incorporating peak limiting, level normalisation, and noise removal. Once a recorded or pre-recorded audio file of the musical element has been edited and processed, it can be forwarded to the backend provided by the system to initiate the matching process. On upload of the audio file (also referred to as an audio loop), the user will designate the file as one of a number of musical categories, dependant on the system calibration. In one embodiment, the user is requested to identify the musical element in the audio file as either a vocal, beat, or melodic file. In other embodiments, the user may be requested to designate the musical element more simply as either vocal or instrumental. The uploaded audio file may also include relevant metadata, for example describing the genre of the musical element, the tempo, and loop length, as required by the system and defined by the user.
The backend provided by the system 20 receives an audio file from the frontend provided by the user device 28 and processes the musical element in the audio file through a uniquely trained audio matching artificial intelligence (Al) algorithm. The Al algorithm is designed to provide intelligently chosen musical elements, from the database 24, to match with the received musical element to create a musical segment.
For example, when the Al algorithm is presented with a beat musical element from the frontend, the Al algorithm intelligently selects a melody and vocal element to combine -13 -with the rhythmic element. Alternatively, when the Al algorithm is presented with a vocal element, it finds a matching rhythmic and melody element.
In some embodiments, the Al algorithm incorporates a number of decision-making processes to identify matching musical elements for any single musical element that is received. In these embodiments, the Al algorithm is configured to first identify metrics and scores related to the received (primary) element's sonic qualities, including, but not limited to, the key, time signature, patterns in pitch variation, and patterns in dynamic (i.e. volume) variation. The Al algorithm then identifies which sonic qualities are required to be present in matched elements. As the final audio playback system is capable of accurately modifying the pitch and tempo of any audio file, the Al algorithm is able to identify elements stored in the database which have compatible pitch and tempo qualities. In some embodiments, the pitch and tempo analyses and modifications are implemented by established digital-signal-processing algorithms which work well within threshold limits (and cause noticeable audio artefacts or distortion degradation if used with more extreme settings). For example, the tempo of any musical element can be changed by, for example, ±20% without a noticeable degradation of audio quality; so a 100 bpm element can be matched with any other musical element within the range of 80-120 bpm. A similar approach may be applied to the musical key in some embodiments, which may also be modified by, for example, five musical semitones without noticeable degradation of audio quality; so an element in the key of C major can be successfully modified to become in the key of D major or E major, amongst others. Additionally, it is known through musical theory that every major key has a relative minor key, so for example a musical element in the key of C major can also be matched with a musical element in the key of A minor. The thresholds for pitch shifting and time stretching can be calibrated and optimised for each audio category. For example, it may be found that rhythmic elements can be time stretched (tempo modified) by 20% with no significant degradation of audio quality, whereas vocal elements are observed to show noticeable sound quality degradation if they are tempo stretched by more than 10%. This applies equally to pitch modifications, where it may be observed that melodic elements can be pitch-shifted to a greater degree than vocal elements before sound quality degradation is noticeable, and threshold settings can be calibrated accordingly. Furthermore, time signatures and rhythmic patterns can be identified and used for judging matching or compatibility between audio elements. For example, musical elements performed in a 4/4 time signature match with other musical elements in a 4/4/ time signature and do not match -14 -with musical elements in a 3/4 time signature. Equally, swing rhythms using triplet time do not match musically with melodic elements that use straight sixteenth note timing. It is therefore possible, by implementing standard knowledge of musical theory and with knowledge of acceptable time and pitch shifting thresholds, to identify from a database of many musical elements, which elements stored in the database will musically match a single received primary element and which will not.
The audio matching by the above method can be enhanced further by evaluating musical elements as combinations of shorter sub-sections of the musical element. For example, an eight-bar musical element can be considered as four consecutive two-bar musical elements or eight consecutive one-bar elements. In some embodiments, the Al algorithm is configured to evaluate the key and tempo in each sub-section, which allows for the identification of potential match elements by a greater degree of granularity, and hence enables the very best potential matches to be identified from a large database of elements. In particular, it is possible to identify the most powerful musical tonic of each two-bar section of an eight-bar piece of audio. For example, an eight-bar loop may have the following two-bar tonic pattern: C-E-C-G. It is therefore evident that another musical phrase with the same C-E-C-G tonic pattern will match very well. However, through application of western musical theory, it is possible to show that there are also many other tonic patterns which musically match with the C-E-C-G tonic pattern. For example, we see that the notes C-E-C-G are all intervals of the C major music scale and therefore it can be anticipated that the pattern C-E-C-G will match well with other patterns made up of notes from the C major scale, for example C-G-D-E. Additionally, the example notes are all intervals of the E minor scale, so it can be anticipated that the pattern C-E-C-G will match well with other patterns made up of notes from the E minor scale, for example C-A-F#-D. Patterns with notes from non-compatible scales will give poorer musical matches, for example pattern C-G-C#-A will not be expected to match well with C-E-C-G owing to the dissonant C# note in the third sub-section. By evaluating pattern components between potential matches, it is possible to identify how many sub-section matches there are between two musical elements. For example, the first sub-section tonic of an instrumental audio element can be evaluated against the first sub-section tonic of a vocal element, and likewise for the second, third and fourth sub-sections, resulting in a total match value between 0-4. Potential matches with 4 matching sub-sections are considered to be strong matches and those with 0 matching sub-elements are considered to be weak matches. -1 -
In some embodiments, the Al algorithm additionally takes the musically matched audio elements (identified as describe above) and implements a second matching process based on machine learning methods. The machine learnt algorithm is taught from many examples of good and bad matches, which are selected by expert music producers, artists and, music industry executives, enabling the Al algorithm to perform an equivalent human selection at a far greater speed and across far greater permutations of a massive database of audio files than are feasible for a human. The Al machine learning incorporates examples from different musical styles, tastes and genres, allowing matches to be tailored to a specific request by the user. For example, the user may request a hip-hop or rock style match when uploading their musical element. Alternatively, the Al algorithm will use historical knowledge of the user's preferences to provide matches that will be appealing to their personal taste.
Once a combination of matched musical elements has been selected, the audio files containing these musical elements are processed through an Audio Playback Processing algorithm 25 to ensure accurate alignment and summation of the elements. All musical elements are modified with standard digitals signal processing algorithms to ensure that each have the same tempo and same musical key. The specific chosen tempo and musical key can be taken as that of the original uploaded musical element, or may be algorithmically chosen to be a suitable value given the characteristics of the matched elements. For example, in some embodiments, the average tempo of three elements may be taken as the tempo of the musical segment. Elements may also be processed with standard audio mixing tools such as dynamic range compression, reverb, echo, and equalisation, to ensure that an impacfful audio mix is heard when the combinations are summed and played audibly as a single audio file. The processed and aligned elements are then forward back to the frontend for review by the user. In some embodiments, the three respective audio files containing the three matched elements are modified and sent back to the frontend and summed in the frontend. In alternative embodiments, the three modified elements are summed (mixed) to a single audio file containing the generated musical element prior to being forwarded to the frontend. The user is notified that a match is available for review, which allows the user to evaluate one or more matched element combinations. Upon listening, the user may choose to accept the match or request changes to the match combination. For example, the user may request a different vocal element to be used in the match, or reject the match altogether. Rejection of the match or one of the combination elements triggers a second matching process and a new match to be reviewed. In some -16 -embodiments, the number of times a user may reject and request new matches is dependent on the system calibration and the particular user's system privileges.
In some embodiments, the system additionally includes a management access interface 29, allowing the database of audio files to be managed, reviewed, edited, and accessed for administration purposes. The management access interface 29 also allows batch upload of audio files in order to populate the database with content. The management access interface 29 therefore also allows quality control and management of the stored audio files and associated metadata.
Embodiments of the present invention have been described with particular reference to the examples illustrated. However, it will be appreciated that variations and modifications may be made to the examples described within the scope of the appending claims.

Claims (14)

  1. -17 -Claims: 1. A system (10) for generating a musical segment, comprising: a database (14) configured to store a plurality of musical elements in a plurality of user defined categories; a processor (12) operably connected to the database (14), configured to analyse a primary musical element of one of the plurality of categories, identify at least one musical element from another of the plurality of categories stored in the database (14) that matches one or more predefined parameters of the primary musical element, and combine the primary musical element with the at least one identified element to generate a musical segment.
  2. 2. A system as claimed in claim 1, wherein the plurality of categories include any of melodic, vocal, and rhythmic.
  3. 3. A system as claimed in claim 1, wherein the processor and the database are operably connected via a communications network (16), and the system is configured to communicate with respective user devices (18a, 18b, 18c) of a plurality of users via the communications network (16).
  4. 4. A system as claimed in any preceding claim, wherein the processor is configured to receive a musical element from a user and store the received element in the database under one of the plurality of categories.
  5. 5. A system as claimed in any preceding claim, wherein the processor is configured to identify the at least one element from another of the plurality of categories by using artificial intelligence to identify the one or more matching predefined parameters.
  6. 6. A system as claimed in claim 5, wherein the processor is configured to, when more than one element from the another of the plurality of categories is identified as having one or more matching predefined parameters, assigning a score to each of the identified elements and using the identified element with the highest score for the generating the musical segment.
  7. -18 - 7. A system as claimed in any preceding claim, wherein the processor is configured to generate the musical segment by aligning the tempo and key of the elements.
  8. 8. A system as claimed in any preceding claim, wherein each musical element in the database is associated with a user, and the processor is configured to send an alert to each user associated with the elements used to generate the musical segment.
  9. 9. A system as claimed in any preceding claim, wherein the primary element is a melodic element, and the at least one defined parameter includes at least one of a tempo pattern, a pitch pattern, and a musical key pattern.
  10. 10. A system as claimed in any preceding claim, wherein each user is associated with a user profile stored on the database.
  11. 11. A system as claimed in any preceding claim, wherein the primary musical element is received from an engaged user.
  12. 12. A system as claimed in claim 11, wherein each user profile includes a set of user preferences, wherein the processor is configured to identify the at least one element from another of the plurality of categories based on the user preferences of the engaged user.
  13. 13. A system as claimed in claim 11 or 12, wherein the processor is configured to receive from the engaged user a user preference score for the generated musical segment, analyse the generated musical segment using artificial intelligence, and update the user preferences based on the analysis and the user preference score.
  14. 14. A method of generating a musical segment, comprising: storing a plurality of musical elements in a plurality of categories in a database (16); using a processor (12), analysing a primary musical element of one of the plurality of categories, identifying at least one musical element stored in the database (16) from another of the plurality of categories that matches one or more predefined parameters of the received musical element, and combining the received musical element with the at least one identified element to generate a musical segment.
GB2106611.3A 2021-05-10 2021-05-10 A system and method for generating a musical segment Withdrawn GB2606522A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
GB2106611.3A GB2606522A (en) 2021-05-10 2021-05-10 A system and method for generating a musical segment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
GB2106611.3A GB2606522A (en) 2021-05-10 2021-05-10 A system and method for generating a musical segment

Publications (1)

Publication Number Publication Date
GB2606522A true GB2606522A (en) 2022-11-16

Family

ID=77021650

Family Applications (1)

Application Number Title Priority Date Filing Date
GB2106611.3A Withdrawn GB2606522A (en) 2021-05-10 2021-05-10 A system and method for generating a musical segment

Country Status (1)

Country Link
GB (1) GB2606522A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210272543A1 (en) * 2020-03-02 2021-09-02 Syntheria F. Moore Computer-implemented method of digital music composition

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100288106A1 (en) * 2006-05-01 2010-11-18 Microsoft Corporation Metadata-based song creation and editing
US20200090632A1 (en) * 2018-09-14 2020-03-19 Bellevue Investments Gmbh & Co. Kgaa Method and system for hybrid ai-based song construction
US20200357371A1 (en) * 2019-05-07 2020-11-12 Bellevue Investments Gmbh & Co. Kgaa Method and system for ai controlled loop based song construction

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100288106A1 (en) * 2006-05-01 2010-11-18 Microsoft Corporation Metadata-based song creation and editing
US20200090632A1 (en) * 2018-09-14 2020-03-19 Bellevue Investments Gmbh & Co. Kgaa Method and system for hybrid ai-based song construction
US20200357371A1 (en) * 2019-05-07 2020-11-12 Bellevue Investments Gmbh & Co. Kgaa Method and system for ai controlled loop based song construction

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210272543A1 (en) * 2020-03-02 2021-09-02 Syntheria F. Moore Computer-implemented method of digital music composition
US11875763B2 (en) * 2020-03-02 2024-01-16 Syntheria F. Moore Computer-implemented method of digital music composition

Similar Documents

Publication Publication Date Title
CN103959372B (en) System and method for providing audio for asked note using presentation cache
CN106023969B (en) Method for applying audio effects to one or more tracks of a music compilation
Frühauf et al. Music on the timing grid: The influence of microtiming on the perceived groove quality of a simple drum pattern performance
US9177540B2 (en) System and method for conforming an audio input to a musical key
US8785760B2 (en) System and method for applying a chain of effects to a musical composition
US8779268B2 (en) System and method for producing a more harmonious musical accompaniment
KR100731761B1 (en) Music production system and method by using internet
CN102760426B (en) Searched for using the such performance data for representing musical sound generation mode
Chau et al. The emotional characteristics and timbre of nonsustaining instrument sounds
Lavengood What makes it sound’80s? The Yamaha DX7 electric piano sound
US20180139268A1 (en) Music selection and organization using audio fingerprints
EP3759706B1 (en) Method, computer program and system for combining audio signals
MX2011012749A (en) System and method of receiving, analyzing, and editing audio to create musical compositions.
CA2929213A1 (en) System and method for enhancing audio, conforming an audio input to a musical key, and creating harmonizing tracks for an audio input
CA2843438A1 (en) System and method for providing audio for a requested note using a render cache
JP2008276169A (en) System and method for online composition, and computer-readable recording medium
GB2606522A (en) A system and method for generating a musical segment
Sethares et al. Expressive timbre and timing in rhythmic performance: analysis of Steve Reich’s Clapping Music
JP7544154B2 (en) Information processing system, electronic musical instrument, information processing method and program
US20240304167A1 (en) Generative music system using rule-based algorithms and ai models
Landry The Electric Marimba: A Study of Selected Works for Marimba with Electronic Effect
JP6611633B2 (en) Karaoke system server
Xue et al. Effective acoustic parameters for automatic classification of performed and synthesized Guzheng music
Manilow Score-Informed and Hierarchical Methods for Computational Musical Scene Analysis
Nieto et al. Android app for automatic music composition

Legal Events

Date Code Title Description
WAP Application withdrawn, taken to be withdrawn or refused ** after publication under section 16(1)