System and method for supporting coherent rhythms for instruments in a generative composition

Abstract

A processor-based automated method of identifying and assembling tessellating rhythmic patterns for a generative composition system maintains coherent musical form between musical bars. The applied method artificially controls, over time, tension and resolution in music to maintain good form by manipulating the correlation between rhythmic output and its rhythmic template. The system takes a time signature for a musical bar, and partitions it into multiple timeslots (TSn) of equal duration. A rhythmic template generator (112) partitions the timeslots into a grid (402) such that combinations of groupings of multiples of two timeslots, three timeslots or combinations of groupings of two and three timeslots are identified within a single bar for the bar. For each grouping, a single rhythmic event marker or "beat" is initially placed in its first timeslot, with beat insertion also completed for all alternative rhythm templates of the grid. Each rhythmic template provides a tessellation over a selectable number of bars. Following selection of one rhythmic template as a prime or root, its event marker positions are populated with notes using probabilistic note position selecting algorithms (619a-d) specific to particular instruments (609, 611, 614, 616), and styles. Since all selected instruments contributing to rhythm share a common root, even with varying note density, musical coherence is observed, and good form maintained between tessellating bars. Harmonics later overlay the generated rhythm(s) in a complementary process.

Description

[0001] SYSTEM AND METHOD FOR SUPPORTING COHERENT RHYTHMS FOR INSTRUMENTS IN A GENERATIVE COMPOSITION

[0002] Background to the Invention

[0003] This invention relates, in general, to signal processing and the automated selection by an apparatus of contextually relevant audio components that can be assembled as a complementary musical overlay to build coherent rhythmic phrasing within a new and original generative composition. More particularly, the invention relates to an automated processor-based system in which rhythmic coherence is based on an automated selection of a relevant skeleton, such as for pitched notes, arranged to realise a musical rhythmic structure linked to a selectable musical genre. The present invention provides a mechanism by which musical good form is maintained, in response to a definition of musical requirements or an explanatory narrative for a musical journey, and which has an output delivered in real-time or otherwise (following streaming or download over a network) from an audio system including a speaker.

[0004] Summary of the Prior Art

[0005] Simply put, music reflects ‘observations’ by the mind that provide a response in the brain.

[0006] Historically, composition has on a sliding scale been a mix between a laboursome task and an inspirational process through which a musician applies their innate understanding of effective musical structures to generate an original composition. If this were not to be the case, the concept of music and particularly its output would degenerate into essentially an entirely random selection of notes placed arbitrarily - or not - into some musical phrasing. Indeed, music theory has traditionally been more of a folk psychology used to name and categorise music, rather than a theory in a scientific sense that can predict the effectiveness of a passage, or the next note or chord in a piece.

[0007] A profound couple of statements but reflective of the fact that music and, more particularly, the appreciation of music reduces to signal processing and mental stimulation associated with the interpretation of a subjectively constructed journey in sound that exploits the concepts of “tension” and “release” as each resolved in the mind of the listener. Regardless of what music amounts to and whether it is based on western, tribal, or oriental structures, there are desirable physiological effects associated with music, with these effects further affecting emotional responsiveness and demeanour.

[0008] There is, however, a requirement to move away from traditional approaches requiring high levels of competence in musical understanding, especially in the context of automated generative composition in which a layperson makes use of processor-based technologies, including interpretational artificial intelligence “Al” systems, to build a multiplicity of orchestrations and musical arrangement that appropriately satisfy musical form requirements. These processor-based systems must be relatively simplistic in execution but must be capable of presenting and retaining the richness and depth in any final arrangement, thereby avoiding the clunkiness of just a succession of discrete notes or chords having little or no rhythmic coherence. Such automated generative composition systems therefore need effectively to prescribe a mechanism of both on-beat and off-beat phrasing, and when a change can be made between the two to support good musical form in auto-generative composition that engages / enhances interest in the listener. On-beat and off-beat changes can be considered as “polarity changes” in emphasis within rhythmic patterns.

[0009] ‘Good’ music - in the sense of an artistically appreciated structured composition - is music that the mind (i.e., relevant neural pathways and centres of the brain) models successfully by being able to predict both an increase in tension within a musical journey and then the following release of that tension. Alternatively, this can be thought of as a compositional piece asking a question, as reflected in musical phrasing or musical structure, and then the compositional piece answering that question [shortly after the question has been posed] to permit mindful termination of a particular part within the entirety that is the musical journey in the composition. The question is thus a construct of tension in the music, and the release a construct that correlates to an appropriate musical answer that puts the change in tonality into perspective. A more complete definition is provided below for these terms to enhance the reader’s understanding of what these semantic terms mean in a more technical sense. Putting the above into a psychological perspective, “good music” is recognised through a self-gratification process in which the mind firstly predicts what it thinks will be delivered by the musical journey, and when an ‘I was right’ prediction is confirmed the reward system of the brain triggers to complete the reward. Whilst not wishing to be bound by theory, it is understood that the reward system refers to a group of structures that are activated by rewarding or reinforcing stimuli. When exposed to a rewarding stimulus (such as good music), the brain responds by increasing the release of the neurotransmitter dopamine. The structures associated with the reward system are found along the major dopamine pathways in the brain, including the ventral tegmental area “VTA” and the nucleus accumbens in the ventral striatum. Another major dopamine pathway, the mesocortical pathway, travels from the VTA to the cerebral cortex and is also considered part of the reward system. This monitorable activity reflects the technical effect that good form produces.

[0010] In contrast, “bad music” or bad composition or “bad form” corresponds to reduced reward / gratification that arises from the brain’s inability to predict anything from seemingly / ostensibly meaningless random [musical] events, and thus the brain’s inability to congratulate itself with a reward arising from stimulation.

[0011] A significant problem that has prevented the effective automated generation of ‘good’ music is “form” and how a rudimentary musical structure can be augmented into a multidimensional arrangement for multiple instruments. The question is how to implement technically a process that does not generate randomness and which technical system is imbued with a technical mechanism that provides selects and then overlays fundamentally compatible musical components such that those musical components are concatenated together seamlessly to provide a new generative composition; this is far from simple.

[0012] In fact, with respect to “form,” composers have previously required experience to identify “form”, and even accomplished composers frequently have failed to appreciate acceptable form until later in their evolutionary compositional life. Even with the gained appreciation of form, human composers frequently revert to templates in all their compositions. Templates provide structure on which the desired narrative is hung. A template can, for example, be sonata form or a rondo and other forms, as will be understood. As a specific example, an underlying style in a first movement of any symphony or concerto will share an identical form but a different narrative, e.g. A-B-A-B-C and then D, where A is the first subject in the major / dominant tonic, B is a contrasting key centre to the major / dominant tonic and A and B together form the “exposition”, C is the conflict between A and B (which is also known as the “development”) and D is the “recapitulation” or resolution of A and B.

[0013] Styles of music will therefore have identifiable characteristics, such as time signatures used to define how an arbitrary duration is sub-divided in a regular manner to provide consistency throughout each section. Conventionally, the sub-divisions of the duration are termed “bars,” with each bar then further divided into an arbitrary number of slots (such as “semiquavers” or “quavers” which are both just measures of time, although other these terms of non-limiting and should be understood to just be labels for conventionally fractional divisions). Consecutive slots are then grouped into beats, with each beat in a given time signature containing the same number of slots. The sub-division into beats provides a structural framework for time in the same way as bars provide a higher-level structural framework for counting time. This conventional time partitioning therefore simply provides a grid which can be filled with rhythmic information realised by sound. A non-random rhythm is therefore nothing more than a defined pattern of sounds that repeat, i.e., tessellate, over a given timeframe (such as a plurality of consecutive bars or otherwise time- separated bars). Sounds may be considered to be any single frequency or multiple simultaneous frequencies, whilst sounds associated with rhythm can nevertheless vary in terms of their frequency components.

[0014] “Form”, in contrast with “narrative” [the latter being what one intends to express musically, i.e., the story between a beginning and end point as expressed by a set of emotional icons such as intensity swells and climaxes], is the structure of linking musical elements together in a musically sensible fashion that avoids discontinuity or randomness (in musical terms) such that a smooth transition is achieved between the syntax of the composite elements. Expressing “form” more tangibly but still subjectively, “good form” may be the syntax reflected in codes and conventions in accepted musical compositions, whereas “bad form” has no obvious or known linking that makes any discernible musical sense between successive musical elements / phrases and, indeed, “bad form” [in music] will fail to communicate structure because the sound signals cannot logically be processed by the brain.

[0015] In order to generate musical good form, the composition - whether autogenerated or not - will be supported by some underlying form function. This could be based on logical operative selection between musical phrasing that is one of a question, an answer, or a statement. This is described, for example, in EP4068273 “SYSTEM AND METHODS FOR AUTOMATICALLY GENERATING A MUSICAL COMPOSITION HAVING AUDIBLY CORRECT FORM,” assigned to DAACI Limited. A question is a chord scheme that suggests tension requiring mental settlement as indicated by notes that have appeared within a harmony or melody and which are questionably present because they are outside of the key centre of the local tonic. Multiple successive questions can be asked musically. An answer is the resolution of the question which operates to resolve the presence of the questionable tones (i.e., pitch) or notes (i.e., pitch with duration) from the mind’s perspective by reinforcing the key centre of either the local tonic or any new tonic of the answering notes. An example of this are the opening two phrases of “The Love Theme” from Superman by John Williams. Finally, a statement is entirely self-contained from a musical question and doesn’t imply or induce any meaningful musical tension that requires release through resolution. A statement is neither a question nor an answer.

[0016] EP4068273 describes, in detail, processes by which complementary notes and chords can be selected to ensure harmonic consistency in the generative composition. The process is based on probabilistic selection of notes and chords. More specifically, EP4068273 describes a generative composition system which makes use of the concept of “Form Atoms.” The system of EP4068273 operates to assemble, automatically, a generative composition having regard to the briefing narrative through selection and concatenation of Form Atoms having tags that align with emotional descriptions timely required by respective ones of the plurality of musical sections; and select and substitute Form Atoms into the generative composition, the substitute Form Atom: derived from the historical corpus of music; and having its compositional heuristics aligned with the emotional descriptions. Form Atoms contain no audio / audio snippets. Form Atoms are generated using their progression and form descriptors, i.e., how a Form Atom with the same descriptor can replace another, regardless of what metadata sits inside the Form Atom regarding stored chords, chord spacer heuristics and alternate chord schemes and rules for their generation. The independent claims articulate this. Form Atoms contain unique data structures that define viable concatenations deliberately selectable by the processing intelligence of the auto-generative compositional system to interpret a briefing narrative.

[0017] The approach in EP4068273 is based on an understanding of composition, and particularly the act of composition, in a conceptually different way, namely: showing how the next note in the audio signals follows an earlier note (as expressed in rules associated with the generation thereof and the length of a fundamental musical component that expresses fundamental audio signal components of a musical section) a rather than what the note actually is. EP4068273 therefore teaches processes by which harmonic melodies are appropriately generated and in which harmonic melodies there is “good form”, thereby addressing the problem in generative composition where there is a need to adapt music to follow a narrative that is different than that laid down by an arbitrary form template regardless of whether that template is human or machine based. In summary, EP4068273 usefully describes how notes and chords can be generated and how adjacent notes or chords are selected to provide harmony. However, the issue of how effective rhythm [with good form in relation to a selectable genre of music] can be generated is another matter entirely, especially in to the context of auto-generative composition.

[0018] Prior to EP4068273, auto-generative systems struggled to realise a generative mechanism that consistently achieved “good form” and thus the generation of relatively high levels of dopamine in the brain’s reward centres. And with a failure to achieve good form and musical depth that sustains interest, by definition the composition acquires “bad form” and correspondingly identifiable qualitative and / or measurable decreases in brain stimulation, particularly associated with the reward centres. Effective generative composition thus leads to a tangible technical effect with an associated technical assessment process. Indeed, better generative composition leads to increasing levels of detectable stimulation / brain activity. In sum, the underlying problems in the process of composition include the requirement for sufficiently maintaining “good form” under the control of the system intelligence assembling the generative work, and particularly supporting variability in the depth of the orchestral arrangement through complementary and coherent phrasing. Putting this differently, how does an auto-generative system for composition decide on the placement of appropriate but variable coherent phrasing to build the orchestration? This problem is particularly prevalent with the requirement to support rhythm in auto-generative composition systems.

[0019] To provide a context for the problem of rapidly generating effective and useful composition(s) in a commercial environment, the generation of an appropriate film score is a first example. Currently, the film director will wite a narrative reflecting the evolution of action in a scene and will then approach a composer for a suitable composition. The composer will review the narrative and attempt to tailor a composition to the narrative in the provision of a “demo” to the client, such as a film director or game designer. This is all time-consuming and requires an innate understanding of musical structure(s). As another example, an interactive game provides no tailored user-experience with respect to the accompanying musical score. Presently, ‘it is what it is’ for the particular aspect of the game or scene in a game and just reflects base programming. Should there be an effective generative process, then the sound experienced in terms of musical textures can provide an enhanced indication for the user as viewed from the emotional perspective of the onscreen avatar. For example, it would be an immersive experience for a player to be exposed to a user-dedicated specific musical segment that reflected growing emotional or physical conditions of the player’ s in-game avatar. Currently, gaming systems provide no audible suggestion of in-game issues that the avatar is facing / experiencing and this is to the detriment of the physical player experience, not least because of the complexity belying the rapid generation of a musically cogent and musically acceptable arrangement for a unique player journey.

[0020] A further application arises with direct augmentation of a musical idea. For example, a person may have a basic melody that requires development. Such development may involve extracting rhythmic data in order to produce complementary melodic structures that develop the original musical idea as a continuing theme. Alternatively, the development may be the production of complementary rhythmic structures in order to generate a more extensive orchestrations for many different instruments and / or styles of music / genres.

[0021] Even prior to the concepts of Form Atoms conceptualised and taught in EP4068273, generative music compilers did exist. These prior existing systems typically used some form of Markov process to generate chords, with all having a series of algorithms that produce different notes across different instruments. The problem with the prior art approaches is that they support little if any creativity and little if any ability to manipulate compositional content. In fact, the prior art approach all generally produce compositions that sound the same because all generated composition is based on a fixed number of predefined instrumental templates. The consequence of this straight-jacketing approach is that a loss of musical texture and orchestrational depth. Moreover, existing systems can be classified as rhythmically naive. This inability to provide, support or produce contextually appropriate rhythm that takes into account user compositional requirements is a significant problem that diminishes usability because of the resultant sameness.

[0022] There are various methods for writing chord schemes that have been implemented over the years (C. Johnson, Carballal, & Correia, 2015; Lerdahl & Jackendoff, 1996; Nierhaus, 2009). The aesthetic valuation for any given method is based on the developer’s artistic requirements, justifications, post-rationalisations, or simple tolerances. Experience in fact shows that it can be considered acceptable for any chord to follow any other chord given enough context in the surrounding harmonic progression. When choosing a chord to follow another one, if this context is ignored and we only look for evidence of the sequence in an example, we find ourselves in the position whereby chord schemes simply become a randomised sequence.

[0023] The following represents a working basis for the terminology used by musicians, and which is relevant to specific embodiments and implementations of the invention: • In Western musical theory, a cadence is a melodic or harmonic configuration that creates a sense of resolution [finality or pause], especially since any cadence has decreasing emphasis. A harmonic cadence is a progression of (at least) two chords that concludes a phrase, section or piece of music. And a rhythmic cadence is a characteristic rhythmic pattern that indicates the end of a phrase. A cadence can be weak or strong depending on its sense of finality. While cadences are usually classified by specific chord or melodic progressions, the use of such progressions does not necessarily constitute a cadence; there must be a sense of closure as at the end of a musical phrase. Generally, harmonic rhythm plays an important part in determining where a cadence occurs. Cadences are also strong indicators of the tonic or central pitch of a passage or piece of music. Currently, harmonic cadences are generally well defined, whereas rhythmic cadence definitions are insufficient since they are assumed to be inherently understood or not. The lack of a technical definition of rhythmic cadence represents a significant problem for auto-generative composition.

[0024] • In music, the tonic is the first scale degree of the diatonic scale (the first note of a scale) and the tonal centre or final resolution tone that is commonly used in the final cadence in tonal (musical key-based) classical music, popular music, and traditional music. In the do solfege system, the tonic note is sung as do. More generally, the tonic is the note upon which all other notes of a piece are hierarchically referenced. Scales are named after their tonics: for instance, the tonic of the C major scale is the note C. The term tonic can also be referred to as a the keycentre. The local tonic, e.g., Cm or Bb, provides both the first and last notes of the scale.

[0025] • A triad formed on the tonic note, the tonic chord, is thus the most significant chord.

[0026] • A chord is a series of pitches played in parallel with each other and which are tied to a keycentre. In terms of function, the mind makes use of a chord to predict where it is in the composition. A chord does not in its own right have any lexicological meaning because musical meaning is derived from the syntax, i.e., the sequence of chords.

[0027] • A chord scheme is a chain of chords.

[0028] • A metachord scheme are the principals of how a chord scheme is written. • Major and minor scales are two of the most popular and commonly used scales in western music, with a set of notes each with a distinct pitch forming the scale. Major and minor scales are variations of the diatonic scale in which there are pitch intervals of five full steps and two half steps, with the relative pitch / physical displacement of the third note determining whether the scale is major or minor. This third note makes the major scale brighter and more cheerful sounding while giving the minor scale its characteristic sadness, melancholy, and darkness. In a major scale, the third note is one note higher than the minor 3rd note. The pattern steps in a major scale has note spacing WWHWWWH (where W representing transition of a whole note and H representing transition of a half note), whereas the pattern in a minor diatonic scale has note spacing WHWWHWW. In convention Western music, any major or minor key will have seven degrees / notes in its scale, i.e., notes A to G.

[0029] Whilst the inventive concepts - of which there are many - will now be described in considerable detail, the following description of additional musical terminology may further assist.

[0030] Particularly in Western music, the relationship between chords is defined by the degree of scale. The degree of scale refers to the position of a particular note (having a particular pitch) on a scale relative to the tonic, i.e., the first and main note of the scale from which each octave is assumed to begin. In music theory, a diatonic scale is any heptatonic scale that includes five whole steps (whole tones) and two half steps (semitones) in each octave, in which the two half steps are separated from each other by either two or three whole steps, depending on their position in the scale. This pattern ensures that, in a diatonic scale spanning more than one octave, all the half steps are maximally separated from each other (i.e. separated by at least two whole steps).

[0031] In overview, tailoring a generative sound experience to a narrative articulated by an end user having no credentials in composition would be advantageous provided that the composition was quickly generated and of a discernible standard and, importantly, includes appropriate emphasis on rhythmic depth for a given orchestration as required to maintain listener engagement. More particularly, to date, the definition of non-random rhythm in generative composition has not been addressed. Consequently, the final quality of auto-generative composition has always been compromised.

[0032] Summary of the Invention

[0033] According to a first aspect of the invention there is provided a processor-based automated method of identifying and assembling tessellating rhythmic patterns for a system supporting rhythmic composition, the method comprising: receiving a duration for a musical bar; sub-dividing the duration into a multiplicity of equal durational timeslots; in a rhythmic template generator, operating within the constraints of the musical bar, partitioning the musical bar into groupings of two adjacent timeslots and groupings of three adjacent timeslots and wherein each of said groupings is comprised from different timeslots and each timeslot is accounted for only once within the duration of the musical bar; for each grouping of two and three timeslots, placing a single rhythmic event marker into a first timeslot of each grouping, thereby producing a grid of alternative rhythmic templates for the musical bar wherein the alternative rhythmic templates are defined by differing positions of respective rhythmic event markers and differing numbers of groupings of two timeslots, three timeslots and integer combinations of two and three timeslots; selecting a first instrument to contribute rhythmically to a selectable genre for a rhythmic composition; selecting from the grid a prime rhythmic template for the rhythmic composition; populating the selected prime rhythmic template for the first instrument with rhythmic components generated using a selectable probabilistic note position selecting algorithm “PNPSA” specific to the first instrument and the selected genre, thereby producing a sensory output supporting rhythmic coherence when the musical bar is tessellated.

[0034] The PNPSA takes into account a function supported by the first instrument.

[0035] Tessellation occurs over multiple musical bars. The multiple musical bars may be adjacent and concatenated or separated by an intermediate bar that reduces the frequency of rhythmic repetition. Probabilistically selected rhythmic components used by the first instrument may correspond partially with the rhythm event markers of the selected prime rhythmic template. Alternatively, there may be total correlation.

[0036] To produce musical tension, a limited number of the probabilistically selected rhythmic components used by the first instrument may be deliberately misaligned with the rhythm event markers of the selected prime rhythmic template. Some of the selected rhythmic components are selectively misaligned relative to the rhythm event markers of the selected prime rhythmic template, wherein the misalignment for a rhythmic component is by no more than two timeslots and most preferably no more than one timeslot relative to a timeslot for the rhythmic components in the prime rhythmic template for the first instrument.

[0037] Partial alignment with the prime rhythmic template may be controlled through selection of a note density for the selectable rhythmic components.

[0038] A preferred method may further comprise: selecting at least a second instrument to contribute rhythmically to a selectable genre for a rhythmic composition; using one of: (a) the selected prime rhythmic template, (b) a complementary derivate rhythm template of the selected prime rhythmic template, and (c) a related rhythmic template selected from the grid and which related rhythmic template shares substantial commonality in its initial placement of rhythmic components within its timeslot relative to those in the prime rhythmic template, to populate rhythmic components for the second instrument, wherein population of rhythmic components into timeslots is controlled by a second selectable probabilistic note position selecting algorithm specific to the second instrument and the selected genre.

[0039] A plurality of instruments contributing to rhythm can share a common root with the prime rhythmic template, thereby observing and maintaining at least a relative high degree if not an absolute coherence and good form between tessellating musical bars. The operation of the selectable probabilistic note position selecting algorithm may be randomised under control of system intelligence or influenced by user input or both.

[0040] Allocated timeslot locations for rhythmic components for a selected rhythmic template can be modified by the probabilistic note position selecting algorithm in response to system intelligence identifying a substitutable pattern within the selected rhythmic template, wherein an alternative pattern of rhythmic components relates to the selected genre and is pre-stored in a database.

[0041] In a particular embodiment of many described, at least a second instrument contributing to rhythm has an uncommon root with the prime rhythmic template, thereby supporting musical tension through partial musical coherency with the first rhythm of the first instrument over tessellating musical bars.

[0042] Selection of an instrument specific rhythmic template from the within the grid is randomised by system intelligence or influenced by user input or both.

[0043] The PNPSA defines logic relating to instrumentation and function relating to at least one of: syncopated comping, sparse polyphonic assembly, and accompanying monophonic line, with probabilistic placement of subsets of notes within the available timeslots of the musical bar a function of both selected note density and the PNPSA for the specific selected instrument.

[0044] In a second aspect of the invention there is provided a method of automatically producing a generative composition for audio output over a local speaker or from a remote speaker, the method comprising: identifying and assembling tessellating rhythmic patterns according to the method of any of claims 1 to 15; and attributing harmonic characteristics to the rhythmic components to produce the generative composition.

[0045] The harmonic characteristics may be derived from a library of Form Atoms and related heuristics that together are arranged to satisfy and produce good form. In another aspect of the present invention there is provided a computer-implemented system for assembling tessellating rhythmic patterns for generative rhythmic composition, the system comprising: an interface for receiving an indication of a time duration associated with a musical bar; system intelligence for sub-dividing the indication of the time duration into a multiplicity of equal durational timeslots (TSn); a rhythmic template generator arranged to operate within the constraints of the musical bar and to partition the musical bar into groupings of two adjacent timeslots and groupings of three adjacent timeslots and wherein each of said groupings is comprised from different timeslots and each timeslot is accounted for only once within the duration of the musical bar, and wherein the rhythmic template generator is further arranged, for each grouping of two and three timeslots, to place a single rhythmic event marker into a first timeslot of each grouping, thereby producing a grid of alternative rhythmic templates for the musical bar wherein the alternative rhythmic templates are defined by differing positions of respective rhythmic event markers and differing numbers of groupings of two timeslots, three timeslots and integer combinations of two and three timeslots and the grid is stored in a database; an instrument selector configured to select stored properties of a first instrument that contribute rhythmically to a selectable and identified genre for a rhythmic composition; a template selector configured to select from the grid a prime rhythmic template for the rhythmic composition; a probabilistic note position selecting algorithm “PNPSA” specific to the first instrument and the selected genre, the PNPSA configured to populate the selected prime rhythmic template for the first instrument with probabilistically selected pre-stored rhythmic components, whereby the system is arranged to produce as an output or to send a file supporting a sensory representation of the generative rhythmic composition.

[0046] The PNPSA is further configured to take into account a function supported by the first instrument.

[0047] Rhythmic components used by the first instrument may correspond partially with the rhythm event markers of the selected prime rhythmic template. At least some of the rhythmic components used by the first instrument may be misaligned by the PNPSA with the rhythm event markers of the selected prime rhythmic template.

[0048] Some of the selected rhythmic components are selectively misaligned relative to the rhythm event markers of the selected prime rhythmic template, and the misalignment for a rhythmic component may be by no more than two timeslots and most preferably no more than one timeslot relative to a timeslot for the rhythmic components in the prime rhythmic template for the first instrument.

[0049] A note density selector is configured to affect and control partially alignment of rhythmic components with rhythmic event markers of the prime rhythmic template.

[0050] In an embodiment, the instrument selector is configured to select stored properties of at least a second instrument arranged to contribute rhythmically to a selectable genre for the rhythmic composition, further comprises: a second PNPSA specific to the second instrument and the selected genre is arranged to populate rhythmic components, usable by the second instrument, into timeslots and the second PNPSA in execution of its function references one of: (a) the selected prime rhythmic template, (b) a complementary derivate rhythm template of the selected prime rhythmic template, and (c) a related rhythmic template selected from the grid and which related rhythmic template shares substantial commonality in its initial placement of rhythmic components within its timeslot relative to those in the prime rhythmic template.

[0051] In another embodiment, system intelligence may be configured to identify a substitutable pattern within the selected rhythmic template, and the PNPSA is arranged to modify allocation of timeslot locations for rhythmic components for a selected rhythmic template in response to identification of the substitutable pattern within the selected rhythmic template and wherein an alternative substitutable patterns of rhythmic components relate to the selected genre and are pre-stored in a database. The selection of an instrument specific rhythmic template from the within the grid can be randomised and the sole function of system intelligence, although user input may also be received via a user interface to affect overall operational control.

[0052] Amongst other functions, the PNPSA can define logic relating to instrumentation and function relating to at least one of: syncopated comping, sparse polyphonic assembly, and accompanying monophonic line, with probabilistic placement of subsets of notes within the available timeslots of the musical bar a function of both selected note density and the PNPSA for the specific selected instrument.

[0053] In yet another aspect of the invention there is provided a processor-based generative composition system arranged automatically to produce a generative composition for direct audio output over a local speaker or to communicate over a network a file of the generative composition to a remote playback device, the generative composition system comprising: the system for assembling tessellating rhythmic patterns according to any of claims 18 to 27 ; and a computer-based system arranged to select from a database and attribute harmonic characteristics to the rhythmic components to produce the generative composition.

[0054] Harmonic characteristics can be derived from a library of Form Atoms and related heuristics.

[0055] Advantageously, the present invention supports the generation of rhythmic coherence in orchestration in an auto-generative composition system through provision of rhythmic templates (or skeletons) for positioning of pitched notes, with each rhythmic templates being a selectable high-level entity which, when selected, determines complementary and coherent rhythmic phrasing prescribed by on-beat and off-beat timing and change opportunities therebetween. Consequently, the present invention realises an audio output in which the coherent musical layers are formed. The present invention provides, in fact, a system and methodology in which rhythmic tension and release can be controlled within a generative composition. More particularly, the system supports an applied method which artificially controls, over time, tension, and resolution in music to maintain good form by manipulating the correlation between rhythmic output and its rhythmic template. The system of the present invention provides a method of generating tessellating rhythms that remain musically coherent in the context of the generated composition. The present invention also generates non-tessellating rhythms that are musically coherent in the context of the generated composition. These can coexist in the generated composition, but that does not need to be the case.

[0056] Coherence is associated with good form in that, within bars based on a common rhythm, an underlying skeleton in that common rhythm is always maintained. This satisfy the issues of tension and release in music, and which is therefore consistent and controlled.

[0057] Beneficially, based on processing music information retrieval techniques and analysis supported by a processor-based system intelligence, such as a bespoke expert system, the present disclosure provides a multiplicity of complementary yet inventively different technical solutions, especially related to the generation of viable and selectable rhythm patterns relevant to generative composition.

[0058] Applications of the techniques of the embodiments and aspects of this disclosure can be employed in any music or video application, including film score, advert production and gaming (especially in the context of producing a user-specific musical accompaniment that is generated to reflects player-selected music having direct player connotation to player emotion(s)). Also, since rhythmic skeletons and resultant generative pieces embody “good form” and originality, the application of the technology can be applied to produce a new composition for which lyrics can be written.

[0059] The present invention produces alternative coherent rhythm templates which are equally satisfiable to the mind from a process that produces usable underlying rhythmic beats that are not reliant upon conventional time signatures and their inherent limitations in generative composition. Once generated and selected (whether automatically or with user input), these rhythms can be populated and enhanced with appropriate musical elements from different musical reference sources. The present invention applies aspects of mathematical function, particularly in the context of time partitioning and related processing of time events, in a way which permits the rapid generation of music possessing good form, with the resulting generative orchestration readily adaptable with different rhythm templates selectable through a software environment.

[0060] The processor-based system of the present invention is relatively simplistic in execution as a computer-implemented invention, retaining richness and depth in any final arrangement thereby avoiding the clunkiness of just a succession of discrete notes or chords having little or no rhythmic coherence. The automated generative composition system effectively prescribes both a mechanism for positioning on-beat and off-beat timings and how a change can be made between the two to support good form / good music required to engage / enhance the interest of the listener. This is not a simple mental process or mere processing, but rather the implementation of a conceptualised technical process based on an assessment of discernible physical properties within groupings of musical pulses, such as timings represented by quavers and semiquavers or any other user-defined time signature.

[0061] The system of the invention and preferred embodiments provides a framework for crafting iterations in generative composition. It offers a way for users to state an intent (in the form of an inputted narrative or brief that is interpreted and correlated to heuristics and thus salient rhythmic sections can be concatenated together in an auditory seamless fashion), and then, indeed, to adjust quickly the output from a briefing specification, such as a beat in an automatically captured reference melody.

[0062] With the present invention, the composers themselves become both the programmers and the users. Composers now use the tool to create the heuristic processes that can be used by other users, thus taking on the technical role of programmers, whereas the commissioners themselves can become composers, as users of the generative tool. The system eliminates the need for expert knowledge in composition.

[0063] Since all selected instruments contributing to rhythm generally always share a common root, even with varying note density, musical coherence is observed, and good form maintained between tessellating bars. Harmonics later overlay the generated rhythm(s) in a complementary process.

[0064] Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings in which:

[0065] FIG.l is a representation of a system for producing a generative composition according to an embodiment of the present invention;

[0066] FIG. 2 is comprised of FIGs. 2a to 3d and shows how groupings of two and three timeslots are formed, according to an embodiment of the present invention, within a grid of equal duration timeslots of a musical bar;

[0067] FIG. 3 is a table reflecting the possible number of groupings of two and three timeslots of the grid of FIG. 2, and within each grouping a single beat is present;

[0068] FIG. 4 illustrates a grid of an eight-timeslot bar partitioned, according to an embodiment of the invention, into groupings containing beats and for which each rhythmic template is paramertised for genre classification;

[0069] FIG. 5 is a representation visualizing a process of known beat identification;

[0070] FIG. 6 reflects an exemplary user interface 600 supporting rhythmic generation according to the present invention;

[0071] FIG. 7 shows a short 7-bar generative rhythmic pattern making use of generative rhythmic templates of a preferred embodiment of the present invention;

[0072] FIG. 8, comprised from FIGs. 8a to 8c, is a flow diagram of the rhythmic development for tessellation over successive bars according to the present invention as a basis for at least a rhythmic pattern of a generative composition supported by preferred embodiments of the present invention.

[0073] FIG. 9, comprised from comparative FIGs. 9a and 9b, showing alternative rhythmic templates for contemporaneous execution as rhythm and counter / complementary rhythm for musical tension development, as supported by embodiments of the present invention; and

[0074] FIG. 10 illustrates how a rhythm template, selected from a grid such as exemplified in FIG. 2, can be further adapted to reflect a further embodiment of the present invention. FIGs. 10 to 17 are a succession of coherent rhythms generated by the system of FIG. 1 having regard to exemplary probabilistic note position selecting algorithms “PNPSA” defined for an instrument style.

[0075] Detailed Description of a Preferred Embodiment

[0076] Generation of a suitably applicable rhythm having good form for a particular musical style is a process related to, but distinct from, the process of finalising the harmonic components applied within any generative composition. The final musical output, e.g., MIDI, is a combination of both in which the first is no more important than the second.

[0077] To this end, FIG.l is a representation of a system 100 for producing a generative composition according to an embodiment of the present invention. More particularly, FIG. 1 shows functional blocks and processor-based components that support generation and use of rhythm templates 102 according to the present invention. The system components are, for purely illustrative purposes, shown generally to reside on a serverside platform instantiated through remote a log-in, although the functionality may equally be downloadable to, or provided as an app, over a network 106 to end users 104. The illustrated components represent interactional functional aspects of what could be a distributed system based in multiple locations, but equally could all be locally stored and locally accessible on a personal computer, smartphone, or the like.

[0078] The present invention generates tessellating rhythms that are musically coherent in the context of the generated composition and, particularly, as maintained within and between bars of a stipulated time signature. The system of the present invention also provides a method of generating non-tessellating rhythms that remain musically coherent in the context of the generated composition. These can coexist in the generated composition, but that does not need to be the case. Coherence is associated with good form in that, within bars based on a common rhythm, an underlying skeleton in that common rhythm is always maintained. This satisfies the issues of tension and release in music, and which is therefore consistent and controlled. In FIG. 1, the system receives a time signature input, typically through a graphic user interface “GUI” 108 or some equivalent user input device. The time signature relates to a time period [in musical terms a “bar”] that is then divided equally into a plurality of timeslots (or sub-divisions) of equal duration. Each bar therefore contains two to more timeslots, and typically at least eight timeslots and more likely up to about sixteen timeslots. Selection of odd numbers of equi-durational timeslots is not precluded, and timeslots in excess of thirty-two are also not precluded. In the latter case, as will become apparent, operational selection of timeslot groupings becomes impractical in the sense that there is no real-world discernible benefit associated with ever finer granularity beyond a certain point. Referring briefly to FIG. 3, this is a table reflecting the possible number of groupings of two and three timeslots of the grid of FIG. 2, and within each grouping a single beat is present / As can be seen from the data, an 8 timeslot bar [equivalent to a quaver count] will support only ten alternative groupings of two and three timeslots, whereas a twelve timeslot bar can support twenty-nine alternatives and a sixteen timeslot bar [equivalent to semiquaver count] can support ninety alternative groupings and therefore ninety independently selectable rhythmic templates. At thirty-two sub-divisions, there are over eight thousand groupings, with this in practical terms likely to be too many highly similar rhythmic patterns have generally imperceivable rhythmic differences. Earger grid sizes, however, can represent concatenated smaller grid sizes, albeit that the duration of the time signature, i.e., the bar, would potentially remain the same and execution of the rhythm corresponding shortened. Larger grid sizes also lead to subtle changes in rhythm templates, with this meaning that more rhythm templates are available for selection and use.

[0079] These timeslots are processed by a grid generator 110 to from a grid.

[0080] The size of the grid both in terms of bar duration and the number of sub-divisions is user selectable since both are arbitrary from the perspective of generating groupings of timeslots into twos and threes.

[0081] Referring again to FIG. 1, a rhythmic template generator 112 operates to produce tessellating groupings of timeslots with a bar, with each grouping comprising two or three [effectively] contiguous timeslots. Referring briefly to FIG. 2, this figure illustrates how groupings of two and three timeslots are achieved. In FIG. 2a, the bar has a duration of simply two timeslots. In FIG. 2b, the bar has a duration of three timeslots. In FIG. 2c, the bar is comprised of four equal timeslots which can be split into two groups of two when appreciating that “alternative 2” requires tessellation with a similar bar and that a second grouping of two is formed by timeslot 4 of the first bar and timeslot 1 of an immediately following concatenated second bar (not shown but implied from a wrap-around with the singularly illustrated bar). In FIG. 2d, a bar of five sub-divisions is shown to comprises five alternative groupings of concatenated two or three sub-divisions. The differing depth of shading in FIG. 2d illustrates how these groupings can be considered, with groupings of two shown darker that groupings of three.

[0082] Within each grouping, the rhythmic template generator 112 inserts a marker, such as a note. The note represents a rhythmic event, i.e., a beat. The frequency components of the beat are somewhat irrelevant for the purposes of setting a rhythm. The beat is simply a marker in time providing a skeleton on which rhythmic generative composition can be hung. Preferably, the beat is placed in the first timeslot of the grouping. The population of beats within each grouping within all available groupings for a given time signature, i.e. within a specific bar, amounts to the generation of set of independently selectable tessellating “rhythmic templates” 102 within any one of which a coherent rhythm can be developed irrespective of the number of timeslots or the duration of bar / time signature.

[0083] There are, clearly, different ways to consider the groupings of two and threes and the relative position of its rhythmic event. For example, considering alternative 3 of FIG. 2d, timeslots 1 to 3 could be viewed as a grouping of three, and timeslots 4 and 5 a grouping of two. This simply means that the position of the rhythmic event is relatively offset and occurs at the end of the respective slots. As indicated above, it is preferable that, for the generation of rhythmic templates, the rhythmic event is inserted - or considered to be inserted - at least initially into the first timeslot of the grouping, as reflected in FIG. 2. Making consistent use of, say, the end timeslot timeslots as alternative would potentially add ambiguity and complexity to patterns development. Use of the first timeslot in each grouping therefore represent and optimal solution. Expressing this differently, the rhythmic template generator 112 partitions the grid into every possible combination of groupings of multiples of two timeslots, groupings of multiples of three times slots or combinations of groupings of two and three timeslots that are each present within the duration of a single bar of defined time. For each grouping (whether two to three timeslots), a single rhythmic event “X” is initially placed in one of its constituent timeslots. This beat insertion is completed for all grouping and for the available different alternative grouping patterns (see Table 3 for the specific numbers of alternatives). Each rhythmic template provides a tessellation over a selectable number of bars. Typically, this tessellation will be between adjacent bar but this need not be the case and complementary rhythms may therefore arise, i.e., be present, over shorter and longer periods.

[0084] The set 102 of rhythmic templates (RTi to RTN) for a specific time signature, as subdivided into a selectable number of timeslots, is preferably stored in a reference rhythmic template database 114. Each alternative rhythmic template within each grid is classified with a descriptive header 116, such as a genre name, e.g. “bebop jazz, “Dixieland jazz,” “neo-swing,” Heavy metal,” “Neoclassical dark wave,” of which there are many.

[0085] Parameters 400 (see FIG. 4) for the occurrence of beats “X” (more generally rhythmic event markers) within each rhythmic template “RTn” in each grid 402 for each time signature are used by the system to define characteristics of rhythm aligned with a corpus of music for a specific genre and so define a filter that is applied to align with descriptive headers 116. The parameters 400 therefore act to classify specific musical styles. As such, for each grid 402, the filters parse the set of rhythmic templates down to appropriate usable rhythms for each genre as cross-referenced by the corresponding descriptive header 116. Each filter for each genre may therefore include one or more rhythmic templates, and a rhythmic template may appear in more than one genre. For example, in FIG. 4, given the evident similarities between parameters for RT2, RTe and RT10, these may all describe rhythms in “swamp blues” [to illustrate yet another potential genre classifier of many]. With larger grids, finer granularity of classification is achievable, so FIG. 4 should be considered as simplistic but illustrative of the approach of the present invention to rhythmic generation and selection.

[0086] Amongst other quantifiable parameters known to a musicologist or musician, parameters for the filters may include:

[0087] • the total number of beats within a bar,

[0088] • whether the bar includes an on-beat start, i.e. whether a beat is populated in the first timeslot of a bar,

[0089] • whether the bar includes an on-beat mid, i.e. whether a beat is populated in first timeslot following immediately after half the total number of timeslots within the bar,

[0090] • whether there has been a polarity change within the bar and how many have occurred within a bar, as occurs with each three-slot grouping within a bar,

[0091] • the number of beats within a frame occurring at the point of an off-beat, i.e., occurring at even numbered timeslots TS2, within the bar.

[0092] The grids and rhythmic templates therefore build an accessible reference library.

[0093] Selection of a specific rhythmic template from the alternatives within a grid of the library may be automated by the system intelligence based on a defined genre, or otherwise determined by user entry such as through keyed entry. This function is represented by template selector 122 (of FIG. 1). The user selection can be based on the user hearing and / or seeing the beat pattern generated as an audible and / or visual output respectively through a speaker or a monitor or both.

[0094] All beats within a selected rhythmic template can be selected for population with note specific qualities for selected instruments. Alternatively, the system permits selection of a subset of available beats for insertion into specific timeslots.

[0095] Selection of specific instruments for the reason of beat population by those selected instruments can be randomised by the system intelligence but is usually the operation of an input “check” by a user into an instrument selector 124. The latter allows for some degree of personalised composition preference by the user. Instruments used for rhythmic beat insertion can include all forms of musical instrument, including but not limited to an exemplary list of guitars, bass, drums, pianos, stringed instruments, percussions instruments, cowbells and others readily appreciated.

[0096] Identification and selection of a relevant rhythm template or subset of rhythm templates from any grid may, for example, make use of a text input entered into the system by the user, with the system (or an ancillary component thereto) extracting, via natural language process “NLP,” a sentiment that ties to the genre. This extracted sentiment can then be tied into genre definitions used in the rhythmic template database 114. A trained neural network “ANN,” as described in Emotional Perception Al Limited’s UK patent application GB1904719.0 explains how a continuous embedding space can be formed to allow recommendation of a semantically close file to a data query, with such an ANN therefore providing a viable approach to implementing a text-driven search tool for rhythm template identification. Another approach to identification of a desired rhythm from the multiplicity of possible rhythm templates makes use of an audio sample in which a user provides the sample and the beat pattern is extracted (using frequency analysis tools readily available in the market, such as AudioBox - bigfug creative software). FIG. 5 is a representation visualizing a process of known beat identification in which beats are aligned by analysis to timeslots TSi-s within a succession of concatenated bars, “Bar 1” and “Bar 2.” In the specific case of FIG. 5, a variation in beat position can be seen to occur between Bar 1 and Bar 2, with beats in Bar 1 occurring in timeslots 1, 4 6 and 8 and in Bar 2 at timeslots 2, 4 and 7. The position of identified beats in the sample can be considered as reflective of a desired rhythm within the new generative composition. Consequently, the beat position derived from the audio sample can be correlated to similar patterns within the rhythm templates in the rhythm template database 114 to identify one of more rhythm templates usable for the new generative composition.

[0097] Further methods of user entry of beat are possible, including but not limited to a tapped beat entered through a keyboard / MIDI interface and analysed by the system intelligence to extract the rhythm. Another approach would be to submit an audio file or reference to an audio file that could then be automatically assessed. Other methods will be appreciated by the skilled addressee.

[0098] The system of the invention may therefore collect together closely related alternative rhythmic templates, and present these as a list linked to identified genre characteristics.

[0099] Genre specific beat specifications for instruments are selected from a multiplicity of precompiled genre files 134-138 for each instrument as held in a genre and instrument database 130. Each genre file 134-138 includes a searchable and selectable label 116 which corresponds to the descriptive header 116 applied to each rhythm template RTN in each rhythm template database 114. The selected rhythm template therefore ties the system’s generative output directly to the corresponding genres in the genre and instrument database 130.

[0100] The generative system of the present invention is configured so that selection, such as through a manual “checking” of an instrument inclusion tab 140 for each specific instrument or through a random but automated expert system tasked with selecting the instrument, denotes that the checked instrument will contribute to the generated rhythm as defined by the selected rhythm template. Multiple instruments can be selected to contribute to rhythm generation, as will be discussed in more detail below and with reference to further drawing figures.

[0101] Each genre file, such as a guitar or a piano or a bass, further includes music compositional data 144 related to a particular genre or structured musical style. This music composition data is based on a defined rule set supporting dynamics for music style as well as harmonic information germane to its related genre. The specific nature of such music composition data needed for beat generation is well-known to the skilled artisan. The music composition data may therefore include note generation data, such as a MIDI code, chord data, pitch data and offset data. It will be appreciated that the rules underlying the music compositional data 144 are agnostic to the rhythm templates. Since the system intelligence is now armed with both a rhythm template and an instrument selection for the corresponding genre, a note position generator 146 is configured to populate the rhythm template with instrument notes [or at least specific instrument engagement triggers]. The actual notes that are to be output require harmonic considerations, such as discussed below in outline and in more detail within the aforementioned European patent application EP4068273 assigned to DAACI Limited. The note position generator 146 is preferably responsive to a note density selector 148. Again, functional performance of the note density selector 148 can be user controlled or randomised by system intelligence to provide variation in rhythmic beat contribution, i.e., “note density”, by one or more than one selected instrument. Selectable variation of the note density is provided to support differing music styles.

[0102] FIG. 6 reflects an exemplary user interface 600 that supports the concepts of rhythm generation using a selected rhythmic template 602 [having sixteen timeslots] combined with the selected overlay of instrument beat contributions 606-610 [or “rhythmic components”] aligned with the selected rhythmic template 602. The instrument beat contributions / can: a) Be a subset of the beats in the selected rhythmic template 602, as is the case for a lower selected note density for the bass 609, with instrument beat contributions having both alignment and misalignment with the beats in the selected rhythmic template 602. The instrument beat contributions of the bass, in FIG. 6, are presented for exemplary reasons as alternatives, with bass rhythm contribution 604 having a higher note density than bass rhythm contribution 606, but both with a lower note density than the rhythmic template 602. b) Correspond exactly with the selected rhythmic template 602 and the beat position of the selected rhythmic template 602, such as shown for the contribution of the guitar 610. c) Include some potentially different offset beats for an instrument relative to the selected rhythmic template 602, such as in the case of the rhythm 608 for the “keys” 608. d) Exceed the number of beats of the rhythmic template 602, such controlled by the note density function and noting that this is not shown in FIG. 6). A scale applied to note density is arbitrary and shown in FIG. 6 to vary between 0.24 and 1.0. For the selected bass 611, FIG. 6 shows that a varying rhythm template for the bass can have, in the case of bass pattern 604, four aligned note positions at timeslots 4, 7 13 and 15 the prime rhythm template 622 whereas for bass pattern 606 there is only a single note alignment.

[0103] Whilst the note density option controls the number of note position that will be populated, the position at which notes are located is controlled by the note position generator (element 146 of Fig.1) applying a probabilistic note position selecting algorithm “PNPSA” 619s- 619d based on a library of formed rules for each instrument. The PNPSA function 619, again selectable from a pulldown library feature 620 (or randomly for a fully automated system) defines rules relating to instrumentation and function, as are known to a musicologist. These rules can reflect, for example, functions such as “syncopated comping,” “sparse polyphonic assembly” and “accompanying monophonic line.” The specific but randomised placement of subsets of notes within the available timeslots of the bar is therefore a function of both selected note density and the PNPSA 619 for the specific selected instrument.

[0104] Continuing with the FIG. 6, the note population for the guitar for its selected PNPSA 619 corresponds exactly to the note placement in the prime rhythm 622. In contrast, the keys rhythm 614 differs again, but nonetheless retains coherence with the prime rhythm template in light of note alignment in timeslots 1, 9, 11 and 13.

[0105] FIG. 6 also reflects that, for a given rhythmic template, options (whether user defined through a pulldown menu 611 in a GUI or automatically selected by the system intelligence) are available to select one or more instruments (such as bass 609, keys 614 and guitar 616).

[0106] FIG. 7 shows a short 7-bar generative rhythmic pattern making use of generative rhythmic templates of a preferred embodiment of the present invention. This diagram illustrates that rhythm pattern RTi from the grid has been selected for a first instrument. RTi may correspond precisely with the prime rhythm template 622 but equally it may be a subset. In bar 2, rhythm pattern RTi is modified through the note density and / or PNPSA functions to generate a subset of note positions producing a first modified instrument rhythm RT'i, although RT'i still firmly based on the rhythm pattern RTi. In bar 3, rhythm pattern RTi returns for the first instrument. In bar 4, first modified instrument rhythm RT'i is played again. In bar 5, rhythm pattern RTi is modified through the note density and / or PNPSA functions to generate a second different subset of note positions producing a second modified instrument rhythm RT''i. In bar 6, rhythm pattern RTi is replayed. Bars 1 to 6 represent a first musical section. In bar 7, a second instrument is selected for a complementary but different second rhythmic template RT2 that is similar to the prime rhythm template 622; this will increase musical tension. Again, over a succession of subsequent bars 2 to 4, this second rhythmic template RT2 is either re-played, i.e. tessellated, or adjusted by a probability variance determined by the PNPSA function in the note position generator 146. FIG. 7 shows, also, that there is a discernible musical section break between a first musical section 640 of bars 1 to 6, comprising rhythmic templates based around RTi and RT2, and a second musical section 642 beginning at bar 7 in which its tessellating rhythmic template RT4 selected from the generated grid for the time signature but within which different tessellating rhythmic template RT4 the note positions are significantly different and musically incoherent. Of course, in practice, the rhythmic template and its tessellations will likely exceed 7-bars [which are merely illustrated for exemplary and explanatory purposes].

[0107] Misalignment of positioned notes, i.e., beats, can be musically justified for reasons of generating tension and release, with tension and depth both arising from greater misalignment and release occurring with the return to a primary, i.e., dominant and first selected, rhythmic template.

[0108] From a preferred operational perspective, a rhythmic event within each grouping of each alternate rhythmic templates will start at the very beginning of the grouping within the initial grid. When generating RTs, a rhythmic event is placed in slot one of every grouping of two to three timeslots because, at this point, their only purpose of the generative process is to provide a consistent reference point, i.e., indicating the start of a grouping. Ignoring the issue of syncopation, embodiments of the present invention consider the parallel selection of structurally similar rhythmic templates. For example, the root or prime rhythmic template can be supplemented within the bar with a second and potentially a third rhythmic template provided that the second and third rhythmic templates are assessed as similar to the root rhythmic template; this is illustrated in FIG.7.

[0109] The assessment of similarity and degrees of correlation can be based on a premise of “removed steps” away from the root template, with a “removed step” being a numeric difference in generally one or more of the rhythmic components within the root rhythm template. Removed steps can be considered to be differences of one, two or three values for (a) total hits (e.g., 3 instead of 4 hits is a separation of one), (b) a note on an onbeat start, (c) a note on an onbeat middle, etc., as well as other musical parameter metrics readily known to the skilled addressee as relating to an approach to genre and / or rhythm classification. “Steps away” assessment may also be based on a correlation approach between different rhythmic templates, rather than an adjustment based on genre-based rhythmic template selection parameters. Consequently, two rhythmic templates that are identical except for one rhythmic event displaced by a single timeslot would be considered one step away and closely correlated and usable from a coherency perspective. Greater timeslot separation leads to less correlation and so increased rhythmic tension within the bar and adjacencies.

[0110] Since a rhythmic root key, i.e., selected prime rhythmic template, is generally maintained over the course of the composition, coherence is achieved. Variations from the root give rise to added tension, which is then released within a section of music, comprising perhaps eight or less bars, through better or absolute alignment with root rhythmic template. Maintenance of coherence through a common rhythmic root but adaptation of note positions avoids “sameness” within the generative composition.

[0111] In all cases, all selected instruments importantly share a common selected rhythmic template even when the note density is dissimilar. This common basis ensures that musical coherence is observed, and the generative composition has the basis for good rhythmic form. At this point, there is in fact no requirement for pitch information. Coherence is further consequential of tessellation of the selected rhythmic template.

[0112] Once a populated rhythmic pattern (element 170 of FIG. 1) has been generated, harmonic attribution is applied to the notes to produce a fully developed generative composition. Harmonic, i.e., pitch data, is applied to each note, including those associated with the rhythm, by a note selector 172. The note selector is a processor-based tool that draws note progression from a database 174 of Form Atoms or other harmonic note generator parameter data. The note selector 172 is therefore probabilistic in its approach to applying appropriate notes, chords, and phrasing, and follows the general teachings of EP4068273 as explained in outline in the section below titled “Harmony Generation as Contrasted with Rhythmic Generation: Applying Harmony using Form Atoms." The output from the note selector 172 can therefore be MIDI and is the generative composition that can be played, downloaded, or streamed, as the case may be, over any connected network 106 to the end user(s) 104.

[0113] FIG. 8, comprised from FIGs. 8a to 8c, is a flow diagram of the rhythmic development for tessellation over successive bars according to the present invention as a basis for at least a rhythmic pattern of a generative composition supported by preferred embodiments of the present invention.

[0114] The generative composition process begins with the definition 802 of the time signature in terms of bar length and its partitioning into equal sub-divisions. From this, the grid of rhythmic templates is formed and stored, with each alternative rhythmic template “RTN” defined by an integer number of groups of two and three sub-division groupings and integer multiples thereof that fill completely the available timeslots within a particular tessellating bar, i.e., stipulated time period. Each grid can be permanently stored and, indeed, generated only once, not least because the resulting patterns do not vary for a defined time signature, although (a) the arbitrary position of each alternative rhythmic template is user dependent whilst the (b) the attributed labels for the nature of the rhythm are generally consistent but nevertheless user definable and thus to a degree it is subjective and arbitrary. The labels are a classification for search and cross-referencing purposes. At 806, the system intelligence populates each grouping with a note position, preferably and consistently as the first timeslot within each timeslot grouping. Selection of at least one rhythm template from the alternatives then follows; this can be via a generated list of closely similar rhythm templates in which there is a high degree of similarity in the number and / or position of music parameters (400 of FIG. 4) within the alternative rhythm templates in the grid for the specific time signature.

[0115] For the generative composition or at least for a backbone to any generative composition, the preferred embodiment can focus on the rhythmic aspects rather than to support and implement harmonic component selection for specific notes or chords and their harmonic adjacencies.

[0116] At step 808, at least a prime rhythm template is selected from the grid of alternatives. Differing secondary rhythmic templates can be selected at this point, but alternatively these can be overlaid later.

[0117] Instrument selection 810 follows in which the instrument or instruments responsible for the rhythm are selected. This is typically a user definition, as explained above, by the system intelligence can generate a listing of instruments based a selected genre and rule definition identifying typical complementary instrument and / or playing styles for that genre.

[0118] As will be appreciated, the execution of process steps may vary from those shown in the exemplary flow of FIG. 8. The process blocks are therefore reflective of the processing considerations required to implement the embodiments of the invention and to produce at least a rhythmic pattern if not a full generative composition possessing good form. Consequently, the decision 812 on whether to add another instrument may occur at this point or another point in the flow, with the selection of another instrument requiring a decision 814 on both a predefined and stored playing style for that new instrument in that genre and, further, whether the instrument follows the selected prime rhythm template or a different rhythm template generally coherent (and preferably entirely coherent) with the prime rhythm template. As indicated above, general coherence raises musical tension that, eventually, will be resolved through a timely return in succeeding bars to the prime rhythm template. If the decision is to add 816 another instrument, then the flow returns to an earlier point of the flow, such as step 810. If the prime rhythm is retained 818 as the base pattern, the either the user or the system intelligence makes a decision 820 on the appropriateness of a note density for each selected instrument. Note density may therefore be changed 822 and defined for each instrument to either align with the prime rhythm template (path 824) or otherwise (path 826) to be a subset including some or all and more rhythmic note positions. Once note density is set, the processing intelligence populates 828 the resulting rhythmic patterns with notes using the PNPSA algorithms as explained above.

[0119] At this point, a bar contains a rhythmic template, which bar can already be tessellated.

[0120] At step 830, optionally, a decision 830 can be made as to whether one or more of the prime and / or secondary or tertiary [selected] rhythm templates contain identifiable rhythm patterns that can, optionally, be substituted out [this will be described in more detail below], or whether rhythm adaptation is required in more general terms of generative composition. If yes 832, a decision 834 is made whether there the rhythmic substitution is relatively minor and retains general coherence with the prime rhythm or whether the change in rhythm is such that a new section of music will result. If the decision leads to a conclusion by the system or user that a new section is evolving 836, then the process resets to an earlier point, such as step 808, where an alternative rhythm from the grid is selected and populated. If the decision from block 834 is that there is no 838 significant rhythmic change [so no new section], then the rhythm generation can be completed 840 and set. Setting of rhythm therefore reflects the bar assembly and tessellation process reflected in FIG. 7. The same point of the process could arise from a negative decision 842 from decision block 830.

[0121] At this point, the generated rhythmic proposition can be tested 842, typically through at least the generation of an audio output if not an audio and accompanying visual output. A decision 844 is made as to whether the generated tessellating rhythmic pattern (in whole or part) is acceptable; this is generally a subjective user assessment. If the rhythmic pattern is assessed to be inadequate 846 or poor, based for example on current PNPSA criteria, then the process can be reset to the selection of rhythm templates 808 or another viable point, such as at note density selection. If the rhythmic pattern is assessed to be acceptable with acceptable good form, then the system can populate the rhythm with harmony / pitch data 850 using a Form Atom approach, as briefly described below but otherwise in more detail in DAACI Limited’s EP patent application EP 4068273 A. Harmony and pitch insertion is supplementary to the underlying principles of the present invention, but clearly relevant to a preferred embodiment in which a fully-fledged generative composition is provided by the system of FIG. 1.

[0122] At block 850, the system intelligence, typically Al-based processing, populates harmonies onto the rhythmic pattern and then tessellates / generates 852 the bars for automated composition. Since the Form Atom approach itself permits the adaption of rhythm by the automated system, whether processor controlled or influenced to some extent by user input, a decision 854 can be made as to whether the present compositional construct works with the proposed rhythms and / or harmonies and is thus acceptable or whether an alternative structure is justified immediately or at a successive bar within the generative composition. In the affirmative 856, rhythmic development can return to any earlier point in the process, such as at step 850 or earlier to take into account potential greater change involving rhythm (at step 834). If the generative composition, including the harmonies, are acceptable 860, then the harmony components are initially set and validated for compositional acceptability (decision 864 and path 866). Given an acceptable rhythmic pattern, harmony change may simply involve selection of alternative instrument harmony components 868. If there is no harmonic adaptation deemed necessary (path 870), an optional further check 871 can be undertaken for a change in instrumentation in the next bar or musical section. If yes 872, the process returns 874 to an appropriate early stage, otherwise 876 an assessment 878 can be made about whether the entirety of the generative composition is complete (in terms of rhythm and harmony). If complete 880, the generative composition can be output 882 as MIDI or otherwise provided to a user, such as over a network streaming or final download. If more composition work is required (path 890), then the process can return to an earlier point, such as at the very beginning of rhythm template selection from the grid or a subsequent point (reflecting a less stringent automated re-working of the generative composition.

[0123] The check mechanisms (such as steps 844 and 864) by which an assessment of suitability of the auto-generated but selected rhythm template and / or applied harmonics to the automated rhythm pattern are subjective. As such, these process steps generally require some direct input reflecting a quality perception. This does not detract from the rapid automated process by which effective rhythm templates are automated developed and populated to provide requisite good form.

[0124] There are additional aspects to rhythmic generation that embodiments of the present invention contemplate, including augmentation and substitution of rhythmic components.

[0125] Referring to FIG. 9, comprised from comparative FIGs. 9a and 9b, showing alternative rhythmic templates for contemporaneous execution as rhythm and counter / complementary rhythm for musical tension development.

[0126] In FIG. 9a, which share a common prime rhythm template 622 with FIG. 6, has exemplary instrument rhythm components for a bass 609, guitar 616, cowbells 902 and keys 614.

[0127] In FIG. 9a, the relative depth of shading is representative of note intensity; this is another variation that can be applied to a rhythm template. For guitar 616, as an example, relatively soft beat / note intensity is applied in timeslots TS3, TSe, TS12 and TS16. The auto generated rhythm for the guitar 616 substantially corresponds with the prime rhythm template, with the exception that there’s also an additional inserted beat at timeslot TS9. The rhythm for the keys 614 in FIG. 9a is simply a subset of the prime rhythm template 622 with note to beat correspondence at timeslots TS4, TS7, TS11 and TS15. The developed rhythm for the cowbells contains three different note strength intensities at timeslots; these being soft at TS3, TS9, TS10, TS14, and TS15, relatively loud at timeslots TS4, TS11 and TSie, and strong / loud at timeslots TSi, TSs, TS7 and TS12. This particular cowbell rhythm, as well as that for guitar, can be made subject to a shuffle / repeat function controlled either by user input or randomly selected by the system intelligence, having appropriate regard to the related PNPSA function. Repeat and shuffle are appropriate for subsequent rhythm generation in later bars.

[0128] In FIG. 9b, a secondary rhythm template 904 has been selected. This differs from the prime rhythmic template by the inclusion of a beat at timeslot TSe and omission of a note / beat at TS7. However, there is significant rhythmic coherent overlap, especially in view of the similarity in instruments and the corresponding selection of PNPSA functions for their playing style within the selected genre, but also non-coherence giving rise to a degree of tension.

[0129] In both FIGs. 9a and 9b, the rhythm for the guitar and cowbells is simply busier, thereby generating more interest through syncopation.

[0130] Rhythmic coherence between different instruments is achieved by the automated system of FIG. 1 for a given prime rhythmic template when a subset of rhythmic hits correspond to an identifiable pattern stored within a library, such as searchable database 102 of FIG. 1 , detailing musicological analysis of hits / spaces for a corpus of music of an identifiable genre. This analysis is relatively trivial for a person skilled in the art, although it could also be automated using beat analysis algorithms as referenced herein.

[0131] The PNPSA function 619, which is effectively a library, is made available to ensure that rhythmic styles for an instrument can be applied to any selected rhythmic template. Provision of the PNPSA function means that melody changes, instrumentation and choice of playing style remains the same across all alternative rhythmic templates and can support directly to fit with every bit of melodic input to the system.

[0132] Referring now to FIG. 10, illustrates how a rhythm template, selected from a grid, can be further adapted to reflect a further embodiment of the present invention. In FIG. 10, a portion of a generated sixteen time shot grid is shown. Specifically, nominal prime rhythm pattern 1002 (#23 of 90 possible rhythm template) has been selected. FIG. 10 also shows how a user interface 1000 can support a soft button 1010 either to download a selected one of the MIDI file or to download the audio for a selected (usually a user selected) beat per minute tempo 1004. A note density selector 148 may also contribute to a final pattern generation, as described above. The PNPSA function, as also earlier described, is reflected in the ability to select PNPSA characteristics of which, in FIG. 10, there are four for a particular style of play within a genre. The PNPSA characteristics can take many forms and reflect analysis of a corpus of music in a genre and the storage of these under a selectable descriptive label within the PNPSA pull down menu of FIG. 6. The table 1012 in FIG. 10 reflects thirteen alternative rhythm patterns that are, having regard to four exemplary PNPSA functions [reflective of corpus note patterns for a genre and style], derivative of the prime rhythmic template 622 but coherent alternatives for that prime rhythmic template 622.

[0133] Merely as examples of how substitution of beats can be actioned, the steps may apply the following rules:

[0134] Step 1: Starting at one of TSi and TS9 for the exemplary sixteen timeslot pattern, identify a block of eight notes having a timeslot population pattern in the prime rhythmic template 622 of Don’t Care “DC” whether beat or rest / silence, Beat “X” and Rest / Silence “RS” in the following form: DC=>DC=>DC=>X=>RS=>RS=>X=>RS, and optionally substitute (i.e. a random 50:50 choice) this pattern for coherent pattern: DC^DC^DC^X^RS^X^RS^X.

[0135] This yields two alternative rhythm patterns shown in FIG. 11 , with one being the prime rhythm template 622 and the other having coherent form based on a change of and to the prime rhythm template 622.

[0136] Step 2: Starting at one of TSi and TS9 for the exemplary sixteen timeslot pattern, identify a block of eight notes having a timeslot population pattern in the prime rhythmic template 622 in the following form: DC^ DC^X^ RS^ X^ RS ^X^ RS, and optionally substitute (i.e. a random 50:50 choice) this pattern for coherent pattern: DC^DC^X^RS^X^X^RS^X. This yields two alternative rhythm patterns shown in FIG. 12, with one being the prime rhythm template 622 and the other having coherent form based on a change of and to the prime rhythm template 622.

[0137] Step 1 and Step 2 can be collectively applied to yield four alternative rhythm patterns shown in FIG. 13 in which the first is the prime rhythm template 622, the second pattern being a coherent form based on a change of and to the prime rhythm template 622 using Step 1, the third being a coherent form based on a change of and to the prime rhythm template 622 using Step 2, and the fourth being a coherent form based on a change of the second pattern having Step 2 applied thereto.

[0138] Step 3 For the exemplary sixteen timeslot pattern, for any note patters containing unchanged note positions arising from the application of step 1 or step 2, replace X => RS = RS with X => RS = wx, where the lower case “wx” represents a beat with a weaker intensity relative to a usual full note on beat X. If selected, Step 3 is always applied. Step 3, as applied to the prime rhythm template 622, generates a single altered rhythmic pattern having coherent form with the prime rhythm template 622. The rhythmic outcome of Step 3 is shown in FIG. 14, with this yielding just a single modified pattern of the prime rhythm template 622.

[0139] Step 1 and Step 2 and Step 3 can also be combined, as shown in FIG. 15 to provide four alternative coherent rhythms for this particular PNPSA, with none being the prime rhythm template 622 although all include the beats of the prime rhythm template 622.

[0140] Step 4: For the exemplary sixteen timeslot pattern, for adjacent pairs of groupings of two timeslots having a beat occurring at an on point of the bar, i.e., at TSi, TS3, TSs, TS7, TS9, TS11, TS13 and TS15, optionally transform a pattern of X=>RS to X^wX but (a) not such that the resulting pattern creates a succession of four consecutive hits within the bar or from tessellation with the next bar, and (b) any such change excludes from the available rhythmic patterns the prime rhythm template 622. This results in a pattern shown in FIG. 16 in which there are sixteen alternative coherent patterns.

[0141] Step 1, Step 2 and Step 4 can also be combined together, with this logically leading to the nineteen alternative patterns illustrated in FIG. 17.

[0142] All of Steps 1 to Step 4 can also be combined, with this generating the thirteen patterns shown in the table of FIG. 10 in which none precisely correspond to the prime rhythm template 622 by virtue of the operation of Step 4 but all of which are coherent with the prime rhythm template 622.

[0143] The following sections relates only to the application of pitch and selected harmonies for good form, and therefore it should be considered as complementary to the present invention in that it supports full generative composition of a final piece of music rather than just auto-generated production and selection of rhythmic components.

[0144] Harmony Generation as Contrasted with Rhythmic Generation: Applying Harmony from Form Atoms to Auto Generated Rhythm

[0145] For completeness, once the rhythmic template of the invention has been formed, appropriate harmony can be applied using the Form Atom and texture approaches described in EP4068273 in which a computer-based auto-generative composition system has an input coupled to receive a briefing narrative describing a musical journey with reference to a plurality of emotional descriptions for a plurality of musical sections along the musical journey. The compositional output of EP4068273 is, in fact, derived from this briefing mechanism from which two requirements for the generative mechanism can be extracted (by, for example, NLP or more structured responses to specific question posed in relation to a selectively definable timeline). The two requirements are that: (i) the mechanism can be briefed by a non-musically skilled individual; (ii) the brief can contain information on the connotations that the commissioner desires at any given point in the composition. In EP4068273, the harmony database comprises a multiplicity of Form Atoms having self- contained constructional properties representative of an historical corpus of music. Each Form Atom has: (i) a generative set of heuristics that support generation of a set of chords in a chord scheme or many different sets of chords in the same or different tonics that achieve the same form function and which thus have similar associated emotional / musical connotations, and heuristics that space out temporally any number of generated chords for any given length of musical time; (ii) a tag that describes compositional heuristics of its respective Form Atom; (iii) a chord list in a local tonic where the chord list defines branching structures giving options for generation of different chords from the local tonic, and (iv) a progression descriptor in combination with a form function that expresses musically one of a question, an answer and a statemen. Each Form Atom thus creates a meta-map of a chord scheme in a musical section, wherein musical transitions between Form Atoms are mapped to identify and then record established transitions between Form Atoms in multiple original scores and such that, within the system, groups exist in which Form Atoms are identified as having similar tags but different constructional properties. Processing intelligence, within the system, responsive to the briefing narrative and coupled to the database, wherein the processing intelligence is arranged to: (a) assemble, automatically, a generative composition having regard to the briefing narrative through selection and concatenation of Form Atoms having tags that align with emotional descriptions timely required by respective ones of the plurality of musical sections; and (b) select and substitute Form Atoms into the generative composition. Each substitute Form Atom is derived from the historical corpus of music and has its compositional heuristics aligned with the emotional descriptions. The processing intelligence is further then arranged to cause output of the auto-generative composition as musical output created from applied heuristics within a texture generator of the generative system, said texture generator arranged automatically to select and apply sequential chord schemes to generate a harmonic palette for orchestration of the auto-generative composition and whereby the musical output is made audible from a speaker receptive of the musical output. The entirety of the system is therefore capable of generating automatically a different generative composition in response to at least a change in the briefing narrative. For texture, EP4068273 - and therefore a particular realisation of an embodiment of the present invention - makes use of processing intelligence arranged to assemble the generative composition having regard to the briefing narrative through selection and concatenation of Form Atoms having tags that align with emotional descriptions timely required by respective ones of the plurality of musical sections. The processing intelligence selects and substitutes Form Atoms into the generative composition, the substitute Form Atom obtained with reference to the historical corpus of music and having its compositional heuristics aligned with the emotional descriptions, and wherein musical transitions between Form Atoms are mapped in the generative composition to reflect pre- established transitions between Form Atoms and such that groups exist in which Form Atoms are identified as having similar tags but different constructional properties. The processing intelligence is further arranged, in generating said generative composition, to select and apply a plurality of musical texture groups assembled from musical instrument components, wherein: a) each of said musical texture groups has an associated tag expressing emotional textural connotation; b) each musical instrument component has musical textural classifiers selected from a set of pre-defined musical textural classifiers and such that: i) different musical instrument components may include a differing subset of pre-defined musical textural classifiers; and ii) each musical instrument component has one of a musical accompaniment attribute and a musical feature attribute, and iii) each musical textural classifier that is present within a musical texture group possesses: aa) either no musical feature attribute or a single musical feature attribute, and bb) any number of musical accompaniment attributes, including no musical accompaniment attribute; and c) different musical texture groups can have a tag having a common emotional textural connotation or a similar emotional textural connotation having an association with the common emotional textural connotation tag, but at the same time different musical texture groups have differing subsets of musical textual classifiers or differing s musical attributes for each musical textural component. In a comprehensive realisation, aspects of the processing system are thus arranged to assemble, as the generative composition, relevant selected Form Atoms in which their respective tags are related to the emotional textural connotation tag of related musical texture groups. In automatically generating a musical composition in which the generative composition contains a plurality of musical texture groups, the method described in EP4068273 comprises: i) assembling musical texture groups from musical instrument components, wherein: a) each of said musical texture groups has an associated tag expressing emotional textural connotation; b) each musical instrument component has musical textural classifiers selected from a set of pre-defined musical textural classifiers and such that: i) different musical instrument components may include a differing subset of pre-defined musical textural classifiers; and ii) each musical instrument component has either a musical accompaniment attribute or a musical feature attribute, and iii) where each musical textural classifier that is present within a musical texture group possesses: aa) either no musical feature attribute or a single musical feature attribute, and bb) any number of musical accompaniment attributes, including no musical accompaniment attribute, and c) different musical texture groups can have a tag having a common emotional textural connotation or a similar emotional textural connotation having an association with the common emotional textural connotation tag, but at the same time different musical texture groups have differing subsets of musical textural classifiers or differing musical attributes for each musical textural component; ii) generating at least one chord scheme to a narrative brief, wherein the chord scheme is based on selecting and assembling Form Atoms and the briefing narrative provides an emotional connotation to a series of events; and iii) applying a texture to the at least one chord scheme to generate the musical composition reflecting the briefing narrative; and musical transitions between Form Atoms are mapped to identify and then record established transitions between Form Atoms in multiple original scores and such that, within the system, groups exist in which Form Atoms are identified as having similar tags but different constructional properties and tags of selected Form Atoms are related to the emotional textural connotation tag of related musical texture groups.

[0146] The generative compositional system of the present invention is, predominantly, a software implemented system that is based on a bespoke expert system running code. The system, as will be understood, therefore includes one or more processors. This system intelligence will call on code stored in memory, and will retrieve, manipulate and return data to and from storage, such as a database or other memory storage. The database may be local to the expert system, but equally it may be remotely located and accessible via a wide area network “WAN” or local area network “LAN” with an appropriate network connection. Equally, the user interface may be a computer or other client device that provide an ability to upload, download and / or stream data and media content to any logically appropriate part of the system for reason of storage (in one or more databases), manipulation and / or output (whether streamed or downloaded or imprinted) as a playable media product, including but not limited to a bespoke user-centric and / or user-selected soundtrack for an interactive game. In short, general components of the underlying system architecture are well-known, although the approach to processing, functionality and generative composition efficiency yields manipulated audio signal data (whether aligned with a film brief of for its own sake and purpose) having improved characteristics and qualities and, more specifically, a processor-based automated method of identifying and assembling tessellating rhythmic patterns in a generative composition system that maintains coherent rhythmic form between successive bars of any time signature. The system provides a significant advance in the field of audio signal processing in the context of, particularly, rhythmic generation providing a skeleton for a generative audio composition.

[0147] Unless specific arrangements are mutually exclusive with one another, the various embodiments described herein can be combined to enhance system functionality and / or to produce complementary functions or system that support the effective identification of user-perceivable similarities and dissimilarities. Such combinations will be readily appreciated by the skilled addressee given the totality of the foregoing description. Likewise, aspects of the preferred embodiments may be implemented in standalone arrangements where more limited functional arrangements are appropriate. Indeed, it will be understood that unless features in the particular preferred embodiments are expressly identified as incompatible with one another or the surrounding context implies that they are mutually exclusive and not readily combinable in a complementary and / or supportive sense, the totality of this disclosure contemplates and envisions that specific features of those complementary embodiments can be selectively combined to provide one or more comprehensive, but slightly different, technical solutions. In terms of the suggested process flows of the accompanying drawings, it may be that these can be varied in terms of the precise points of execution for steps within the process so long as the overall effect or re-ordering achieves the same objective end results or important intermediate results that allow advancement to the next logical step. The flow processes are therefore logical in nature rather than an absolute and rigid sequential process and identify both direct technical functional requirements presented in the context of some user interaction(s). The functional architectures of the drawings may be implemented independently of one another, as will be understood, so that the resulting system is a distributed system potentially dispersed via a wide area network, such as the internet. Architecturally, realization of aspects of the system can be implemented using technologies such as the Java Expert System Shell “JESS” and, more typically, a bespoke expert system.

[0148] Aspects of the present invention may be provided in a downloadable form or otherwise on a computer readable medium, such as a CD ROM, that contains program code that, when instantiated, executes the link embedding functionality at a webserver or the like.

[0149] The present invention and particularly the preferred approach to rhythmic template generation is applicable to any musical scale and any cultural precondition, not just Western music.

[0150] From a composition perspective, the identification and generation of rhythmic patterns as a skeleton is highly significant to any generative composition. The complementary classification and manipulation of textures for harmonic overlay, based on the use of Form Atoms, is ancillary but important to any final generative work.

[0151] It will, of course, be appreciated that the above description has been given by way of example only and that modifications in detail may be made within the scope of the present invention. For example, whilst the generative system has been expressed in the context of Western music having a particular degree of scale, the techniques are commutable to other styles and metres.

[0152] The skeletal and analysis techniques, provided with the described generative framework, give a foundation for looking at music hierarchically in a way that leads to effective output. This is not only a useful method of creating aesthetically functional generative film composition and game scores that can, in fact, be orchestrated personally by the user provided that they are given access to the system of FIG. 1.

[0153] Completely autonomous solutions are feasible, based on the given hierarchy, in which computers analyse works and compose music based on analysis. For example, a trained artificial intelligence mechanism, such as deep learning neural networks and generative algorithms with associated fitness functions, to learn how to select appropriate rhythmic patterns.

[0154] Moreover, various aspects or features described herein can be implemented as a method, apparatus, or article of manufacture using standard programming and / or engineering techniques. The term “article of manufacture” as used herein is intended to encompass both a computer program accessible from any computer-readable device or media and also a trained artificial neural network “ANN”. For example, computer-readable media can include but are not limited to magnetic storage devices (e.g., hard disk, floppy disk, magnetic strips, etc.), optical disks (e.g., compact disk (CD), digital versatile disk (DVD), etc.), smart cards, and flash memory devices (e.g., Erasable Programmable Read Only Memory (EPROM), card, stick, key drive, etc.) A trained ANN may be delivered as code or embedded in a processor. Additionally, various storage media described herein can represent one or more devices and / or other computer-readable media for storing information.

[0155] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, methods and algorithms described in connection with the examples disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, methods and algorithms have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application while remaining, either literally or equivalently, within the scope of the accompanying claims.

[0156] As used in this application, the terms “component,” “module,” “system,” and the like are intended to refer to a computer-related entity, either hardware, firmware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to being, a computer-implemented process running on a processor, a processor, an object, an executable, a thread of execution, a program, and / or a computer. By way of illustration, both an application running on a computing device and the computing device can be a component. One or more components can reside within a process and / or thread of execution and a component can be localized on one computer and / or distributed between two or more computers. In addition, these components can execute from various computer readable media having various data structures stored thereon. The components can communicate by way of local and / or remote processes such as in accordance with a signal having one or more data packets (e.g., data from one component interacting with another component in a local system, distributed system, and / or across a network such as the Internet with other systems by way of the signal).

Claims

Claims1. A processor-based automated method of identifying and assembling tessellating rhythmic patterns for a system supporting rhythmic composition, the method comprising: receiving a duration for a musical bar; sub-dividing the duration into a multiplicity of equal durational timeslots (TSn); in a rhythmic template generator (112), operating within the constraints of the musical bar, partitioning the musical bar into groupings of two adjacent timeslots and groupings of three adjacent timeslots and wherein each of said groupings is comprised from different timeslots and each timeslot is accounted for only once within the duration of the musical bar, and for each grouping of two and three timeslots, placing a single rhythmic event marker into a first timeslot of each grouping, thereby producing a grid (402) of alternative rhythmic templates for the musical bar wherein the alternative rhythmic templates are defined by differing positions of respective rhythmic event markers and differing numbers of groupings of two timeslots, three timeslots and integer combinations of two and three timeslots; selecting (810) a first instrument (609, 611, 614, 616) to contribute rhythmically to a selectable genre for a rhythmic composition; selecting from the grid a prime rhythmic template (622) for the rhythmic composition; populating (806) the selected prime rhythmic template for the first instrument with rhythmic components generated using a selectable probabilistic note position selecting algorithm “PNPSA” (619a-619d) specific to the first instrument (609, 611, 614, 616) and the selected genre (116), thereby producing a sensory output supporting rhythmic coherence when the musical bar is tessellated.

2. The processor-based automated method according to claim 1, wherein the probabilistic note position selecting algorithm (619a-619d) takes into account a function supported by the first instrument.

3. The processor-based automated method according to claim 1 or 2, wherein tessellation occurs over multiple musical bars.

4. The processor-based automated method according to claim 3, wherein the multiple musical bars are adjacent and concatenated.

5. The processor-based automated method according to any preceding claim, wherein probabilistically selected rhythmic components used by the first instrument correspond partially with the rhythm event markers of the selected prime rhythmic template.

6. The processor-based automated method according to any preceding claim, wherein at least some of the probabilistically selected rhythmic components used by the first instrument misalign with the rhythm event markers of the selected prime rhythmic template.

7. The processor-based automated method according to claim 6, wherein some of the selected rhythmic components are selectively misaligned relative to the rhythm event markers of the selected prime rhythmic template, wherein the misalignment for a rhythmic component is by no more than two timeslots and most preferably no more than one timeslot relative to a timeslot for the rhythmic components in the prime rhythmic template for the first instrument.

8. The processor-based automated method according to claim. 5, 6 or 7, wherein the partial alignment with the prime rhythmic template is controlled by selection of a note density for selectable rhythmic components.

9. The processor-based automated method according to any preceding claim, wherein the method further comprises: selecting at least a second instrument to contribute rhythmically to a selectable genre for a rhythmic composition; using one of:(a) the selected prime rhythmic template,(b) a complementary derivate rhythm template of the selected prime rhythmic template, and(c) a related rhythmic template selected from the grid and which related rhythmic template shares substantial commonality in its initial placement of rhythmic components within its timeslot relative to those in the prime rhythmic template, to populate rhythmic components for the second instrument, wherein population of rhythmic components into timeslots is controlled by a second selectable probabilistic note position selecting algorithm (619a-619d) specific to the second instrument (609, 611, 614, 616) and the selected genre (116).

10. The processor-based automated method according to claim 9, wherein a plurality of instruments contributing to rhythm share a common root with the prime rhythmic template, thereby to observe and maintain a degree of coherence and good form between tessellating musical bars.

11. The processor-based automated method according to any preceding claim, wherein operation of the selectable probabilistic note position selecting algorithm is randomised under control of system intelligence.

12. The processor-based automated method according to any preceding claim, wherein allocated timeslot locations for rhythmic components for a selected rhythmic template can be modified by the probabilistic note position selecting algorithm (619a-619d) in response to system intelligence identifying a substitutable pattern within the selected rhythmic template, wherein an alternative pattern of rhythmic components relates to the selected genre and is pre-stored in a database.

13. The processor-based automated method according to claim 9, wherein at least a second instrument contributing to rhythm has an uncommon root with the prime rhythmic template, thereby supporting musical tension through partial musical coherency with the first rhythm of the first instrument over tessellating musical bars.

14. The processor-based automated method according to claim 9, wherein selection of an instrument specific rhythmic template from the within the grid is randomised by system intelligence.

15. The processor-based automated method according to any preceding claim, wherein the PNPSA defines logic relating to instrumentation and function relating to at least one of: syncopated comping, sparse polyphonic assembly, and accompanying monophonic line, with probabilistic placement of subsets of notes within the available timeslots of the musical bar a function of both selected note density and the PNPSA for the specific selected instrument.

16. A method of automatically producing a generative composition for audio output over a local speaker or from a remote speaker, the method comprising: identifying and assembling tessellating rhythmic patterns according to the method of any of claims 1 to 15; and attributing harmonic characteristics to the rhythmic components to produce the generative composition.

17. The method of claim 16, wherein the harmonic characteristics are derived from a library of Form Atoms and related heuristics that together satisfy good form.

18. A computer-implemented system for assembling tessellating rhythmic patterns for generative rhythmic composition, the system comprising: an interface for receiving an indication of a time duration associated with a musical bar; system intelligence for sub-dividing the indication of the time duration into a multiplicity of equal durational timeslots (TSn); a rhythmic template generator (112) arranged to operate within the constraints of the musical bar and to partition the musical bar into groupings of two adjacent timeslotsand groupings of three adjacent timeslots and wherein each of said groupings is comprised from different timeslots and each timeslot is accounted for only once within the duration of the musical bar, and wherein the rhythmic template generator is further arranged, for each grouping of two and three timeslots, to place a single rhythmic event marker into a first timeslot of each grouping, thereby producing a grid (402) of alternative rhythmic templates for the musical bar wherein the alternative rhythmic templates are defined by differing positions of respective rhythmic event markers and differing numbers of groupings of two timeslots, three timeslots and integer combinations of two and three timeslots and the grid is stored in a database; an instrument selector configured to select (810) stored properties of a first instrument (609, 611, 614, 616) that contribute rhythmically to a selectable and identified genre for a rhythmic composition; a template selector configured to select from the grid a prime rhythmic template (622) for the rhythmic composition; a probabilistic note position selecting algorithm “PNPSA” (619a-619d) specific to the first instrument (609, 611, 614, 616) and the selected genre (116), the PNPSA configured to populate (806) the selected prime rhythmic template for the first instrument with probabilistically selected pre-stored rhythmic components, whereby the system is arranged to produce as an output or to send a file supporting a sensory representation of the generative rhythmic composition.

19. The system of claim 18, wherein the PNPSA (619a-619d) is configured to take into account a function supported by the first instrument.

20. The system of claim 18 or 19, wherein rhythmic components used by the first instrument correspond partially with the rhythm event markers of the selected prime rhythmic template.

21. The system of claim 18, 19 or 20, wherein at least some of the rhythmic components used by the first instrument are misaligned by the PNPSA with the rhythm event markers of the selected prime rhythmic template.

22. The system of claim 21, wherein: some of the selected rhythmic components are selectively misaligned relative to the rhythm event markers of the selected prime rhythmic template, and the misalignment for a rhythmic component is by no more than two timeslots and most preferably no more than one timeslot relative to a timeslot for the rhythmic components in the prime rhythmic template for the first instrument.

23. The system of any of claims 18 to 22, further comprising: a note density selector is configured to affect and control partially alignment of rhythmic components with rhythmic event markers of the prime rhythmic template.

24. The system of any of claims 18 to 23, wherein the instrument selector is configured to select stored properties of at least a second instrument arranged to contribute rhythmically to a selectable genre for the rhythmic composition, further comprising: a second PNPSA specific to the second instrument (609, 611, 614, 616) and the selected genre (116) is arranged to populate rhythmic components, usable by the second instrument, into timeslots and the second PNPSA in execution of its function references one of:(a) the selected prime rhythmic template,(b) a complementary derivate rhythm template of the selected prime rhythmic template, and(c) a related rhythmic template selected from the grid and which related rhythmic template shares substantial commonality in its initial placement of rhythmic components within its timeslot relative to those in the prime rhythmic template.

25. The system of any of claims 18 to 24, further comprising: system intelligence configured to identify a substitutable pattern within the selected rhythmic template, and the PNPSA is arranged to modify allocation of timeslot locations for rhythmic components for a selected rhythmic template in response to identification of the substitutable pattern within the selected rhythmic template and wherein an alternativesubstitutable patterns of rhythmic components relate to the selected genre and are prestored in a database.

26. The system of any of claims 18 to 25, wherein selection of an instrument specific rhythmic template from the within the grid is randomised by system intelligence.

27. The system of any of claims 18 to 25, wherein the PNPSA defines logic relating to instrumentation and function relating to at least one of: syncopated comping, sparse polyphonic assembly, and accompanying monophonic line, with probabilistic placement of subsets of notes within the available timeslots of the musical bar a function of both selected note density and the PNPSA for the specific selected instrument.

28. A processor-based generative composition system arranged automatically to produce a generative composition for direct audio output over a local speaker or to communicate over a network a file of the generative composition to a remote playback device, the generative composition system comprising: the system for assembling tessellating rhythmic patterns according to any of claims 18 to 27 ; and a computer-based system arranged to select from a database and attribute harmonic characteristics to the rhythmic components to produce the generative composition.

29. The processor-based generative composition system according to claim 28, wherein the harmonic characteristics are derived from a library of Form Atoms and related heuristics.

Images