AI Music Composition Style Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI music composition methods struggle to create new and original musical styles, often replicating past styles rather than reflecting the individual compositional style of the user, due to reliance on large training data sets that overshadow the user's unique style.
Innovation Solution
A method using deep learning AI that generates training data primarily from the user's 'seed' compositions, applying variation and mash-up techniques to create a large dataset that reflects the user's distinctive style, allowing for the generation of novel musical styles by varying and combining seed compositions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If large training data sets are used for AI music composition, then the AI can learn diverse musical patterns and styles, but the user's individual compositional style is overshadowed and lost
Solution Approach 1:
The patent segments the training data into two distinct components: a large public dataset for learning general musical patterns and a smaller private dataset containing user-specific compositions for preserving individual style. This segmentation allows the AI to learn from both sources without the user's style being overwhelmed by the larger dataset.
Solution Approach 2:
The patent applies preliminary action by first training the AI on the user's specific compositions to establish their individual style, then fine-tuning with the larger public dataset. This sequential approach ensures the user's style is established before being exposed to broader musical influences.
2Adaptability or versatility
If AI systems are trained on diverse music data, then they can generate varied musical output, but they fail to create new and original musical styles
Solution Approach 1:
The patent merges two training datasets with different characteristics: a large public dataset providing musical diversity and a smaller user-specific dataset providing stylistic uniqueness. This combination allows the AI to generate varied music while maintaining originality through the user's distinctive compositional patterns.
Solution Approach 2:
The training data is structured as a composite of two components: public music data serving as the base material for general musical knowledge and user compositions serving as the distinctive layer that imparts originality. This composite structure enables the AI to produce music that is both varied and original.
3Stability of the object's composition
If AI composition methods replicate past styles, then they maintain consistency with established music theory, but they cannot reflect the user's unique compositional style
Solution Approach 1:
The patent implements dynamics by making the training process adaptable: the AI can adjust the weight and influence of different training datasets based on the user's needs. This allows the system to maintain music theory consistency from public datasets while dynamically incorporating user-specific stylistic elements.
Data Source
AI summary
A method of creating AI-composed music having a style that reflects and/or augments the personal style of a user includes selecting and/or composing one or more seed compositions; applying variation and/or mash-up methods to the seed compositions to create training data; training an AI using the training data; and causing the AI to compose novel musical compositions. The variation methods can include methods described in the inventor's previous patents. A novel method of creating mash-ups is described herein. Variation and mash-up methods can further be applied to the A compositions. The AI compositions, and/or variations and/or mash-ups thereof, can be added to the training data for re-training of the AI. The disclosed mash-up method includes parsing the seed compositions into sequences of elements, which can be of equal length, beat-matching the seed compositions to make corresponding elements of equal beat length, and combining the elements to form a mash-up.