...

/

Getting Started with Music Generation

Getting Started with Music Generation

Explore some foundational concepts related to music generation.

Audio generation has a number of applications. The most prominent and popular ones nowadays are a series of smart assistants (Google Assistant, Apple Siri, Amazon Alexa, and so on). These virtual assistants not only try to understand natural language queries but also respond in a very human-like voice. Audio generation also finds applications in the field of assistive technologies, where text-to-speech engines are used for reading content onscreen for people with reading disabilities.

Press + to interact
Digital smart assistants
Digital smart assistants

Leveraging such technologies in the field of music generation is increasingly being explored as well. The acquisition of AI-based royalty-free music generation service Jukedeck by ByteDance (the parent company of social network TikTok) speaks volumes for this field’s potential value and impactButcher, Mike. 2019. “It Looks like TikTok Has Acquired Jukedeck, a Pioneering Music AI UK Startup.” TechCrunch. July 23, 2019. https://techcrunch.com/2019/07/23/it-looks-like-titok-has-acquired-jukedeck-a-pioneering-music-ai-uk-startup/..

In fact, AI-based music generation is a growing trend with a number of competing solutions and research work. Commercial offerings for computer-aided music generation, such as Apple’s GarageBand“GarageBand for Mac.” 2020. Apple. 2020. https://www.apple.com/mac/garageband/., provide a number of easy-to-use interfaces for novices to compose high-quality music tracks with just a few clicks. Researchers on Google’s project Magenta“Magenta.” n.d. Make Music and Art Using Machine Learning. Magenta. https://magenta.tensorflow.org. are pushing the boundaries of music generation to new limits by experimenting with different technologies, tools, and research projects to enable people with little to no knowledge of such complex topics to generate impressive pieces of music on their own.

Music generation is an inherently complex and difficult task. Doing so with the help of algorithms (machine learning or otherwise) is even more challenging. Nevertheless, music generation is an interesting area of research with a number of open problems and fascinating works. In this section, we’ll build a high-level understanding of this domain and explore a few important and foundational concepts.

Components of music

...