Schenk Composer: A framework for algorithmic music generation

Unmet Need

Music composition remains largely inaccessible to individuals without formal training in music theory, despite growing interest in AI-assisted creative tools. Existing AI-generated music solutions frequently produce results that are uninterpretable and unpredictable, offering limited control over the creative process and often failing to capture the emotional intent of a composition. Current procedural generation methods operate as black boxes, generating music in ways that are difficult to understand or modify. There is a need for an AI-driven tool that enables both novice and experienced users to intuitively generate expressive, emotion-driven music while maintaining interpretability and control over the compositional process.

Technology

Duke inventors have developed an algorithm to generate musical scores in various musical styles using hierarchical music theory such as form and Schenkerian analysis. This allows the breakdown of music into several fundamental technical components, which can be automated or easily controlled by the consumer. The technology is intended to be provided direct to consumers through a user-interfacing software package that allows one to modify parameters of the algorithm and visualize the results of the system. Specifically, the algorithm employs probabilistic context free grammars and Markov models to analyze musical scores and probabilistically produce new music of a chosen style. This has been demonstrated with a web accessible software platform that allows users to intuitively interface with the algorithm to achieve different musical results. In user testing, the platform and algorithm were able to produce AI-generated music that was perceived as nearly indiscernible from human composed works.

Advantages

• Advanced-stage prototype software already developed
• Algorithm results are fully interpretable; parameter contributions are easily traced to generated musical scores
• Optimized production quality that tests well based off human perception