Automated midi music composition server

1. A music composition system for composing music segments, the music composition system comprising:

an Application Programming Interface (API), wherein the API is configured to process both an internal job request instigated by a component of the music composition system and an external job request originated outside of the music composition system, wherein the API comprises a request manager configured to assign a job identifier to a request for a musical composition;

a controller configured to determine based on the request a plurality of musical parts for the musical composition;

a composition engine configured to generate, for each of the determined musical parts, at least one musical segment in digital musical notation format, wherein the composition engine comprises a plurality of Probabilistic Sequence Models (PSMs), the plurality of PSMs are configured to generate a sequence of musical segments based on probability distribution, the plurality of PSMs each are configured to be specialized for particular musical styles and are associated with a respective style indicator, and the plurality of PSMs are further configured to be coordinated by providing output of one PSM among the plurality of PSMs as input of another PSM among the plurality of PSMs for generating musically cooperating musical segments; and

wherein the at least one generated music segment is stored in association with the assigned job identifier for reproduction at a later time.

2. The music composition system according to claim 1 , wherein the controller is configured to determine from the request a style parameter identifying a style attribute for the composition and the composition engine is configured to generate the musical segments based on the style parameter.

3. The music composition system according to claim 2 , wherein the one or more of the composition modules are based on the style parameter.

4. The music composition system according to claim 3 , wherein each of the composition modules is associated with a respective style indicator, and the one or more composition modules are selected by matching the style parameter to their style indicator(s).

5. The music composition system according to claim 1 , wherein the composition engine has a plurality of composition modules, wherein the composition engine is configured to select one or more of the composition modules based on the request.

6. The music composition system according to claim 5 , wherein each of the composition modules is in the form of the PSM.

7. The music composition system according to claim 6 , wherein the probabilistic sequence model is a trained neural network.

8. The music composition system according to claim 1 , wherein the composition engine has a first composition module configured to generate at least a first music segment for a first of the musical parts, and a second composition module configured to generate at least a second music segment for a second of the musical parts based on data of the first music segment.

9. The music composition system according to claim 8 , wherein the first music segment embodies a chord sequence and the second music segment embodies a melody or harmony composed to fit the chord sequence.

10. The music composition system according to claim 8 , wherein the first music segment embodies a melody and the second music segment embodies a harmonization of the melody.

11. The music composition system according to claim 1 , wherein the composition engine has an input configured to receive a percussion part and at least one of the segments is composed based on the percussion part.

12. The music composition system according to claim 11 , wherein the controller is configured to determine a set of weightings based on the request and the composition settings are selected probabilistically according to the weightings.

13. The music composition system according to claim 12 , wherein the weightings are determined based on the style parameter.

14. The music composition system according to claim 12 , wherein the weightings are determined based on the style parameter.

15. The music composition system according to claim 1 , wherein the controller is configured to select from a set of available composition parameters composition parameters for the composition of the segments using a probabilistic selection process.

16. The music composition system according to claim 1 , wherein the composition engine is configured to store each of the music segments in a database is association with a segment identifier.

17. The music composition system according to claim 16 , wherein the job identifier assigned to the request is associated with each of the segment identifiers.

18. The music composition system according to claim 17 , wherein the response to the request comprises the job identifier and thereby indicates the music segments.

19. A method of composing music segments, comprising:

receiving from an external device a request for a musical composition via an Application Programming Interface (API), wherein the API is configured to process both an internal job request instigated by a component of the music composition system and an external job request originated outside of the music composition system, and wherein the API comprises a request manager configured to assign a job identifier to the request for the musical composition;

determining based on the request a plurality of musical parts for the musical composition;

generating, by a composition engine, for each of the determined musical parts, at least one musical segment in digital musical notation format, wherein the composition engine comprises a plurality of Probabilistic Sequence Models (PSMs), the plurality of PSMs are configured to generate a sequence of musical segments based on probability distribution, the plurality of PSMs each are configured to be specialized for particular musical styles and are associated with a respective style indicator, and the plurality of PSMs are further configured to be coordinated by providing output of one PSM among the plurality of PSMs as input of another PSM among the plurality of PSMs for generating musically cooperating musical segments;

outputting a response to the request; and

storing the at least one generated music segment in association with the assigned job identifier for reproduction at a later time.

20. A non-transitory computer program product comprising executable instructions stored on a non-transitory computer-readable storage medium and configured, when executed at a music composition system, to cause the music composition system to implement operations comprising:

receiving a request for a musical composition via an Application Programming Interface (API), wherein the API is configured to process both an internal job request instigated by a component of the music composition system and an external job request originated outside of the music composition system, and wherein the API comprises a request manager configured to assign a job identifier to the request for the musical composition;

determining based on the request a plurality of musical parts for the musical composition;

generating, by a composition engine, for each of the determined musical parts, at least one musical segment in digital musical notation format, wherein the composition engine comprises a plurality of Probabilistic Sequence Models (PSMs), the plurality of PSMs are configured to generate a sequence of musical segments based on probability distribution, the plurality of PSMs each are configured to be specialized for particular musical styles and are associated with a respective style indicator, and the plurality of PSMs are further configured to be coordinated by providing output of one PSM among the plurality of PSMs as input of another PSM among the plurality of PSMs for generating musically cooperating musical segments;

outputting a response to the request; and

storing the at least one generated music segment in association with the assigned job identifier for reproduction at a later time.