Functional music within the age of AI
Around ten years ago, music streaming services were competing heavily for the very best music advice system. Clearly, a flawless advice system would supply the user with the precise piece of music that optimally satisfies their needs, each time. Nevertheless, some people view advice systems as transitional technology. Ultimately, irrespective of the dimensions of your music catalog, there can’t be an ideal fit available for every possible user request.
Modern generative AI systems could potentially solve this problem by generating music that’s (robot) hand-tailored to every request. Then again, these generative systems are still not producing high-quality music, have tremendous computational costs, and are subject to complex ethical and legal concerns.
Due to this fact, this text goals to check the advantages & limitations of search-based music retrieval and music generation to search out out whether we should always expect generative systems to completely replace, augment, or not even affect the present solutions. Before we start, let’s define what we mean by a “search algorithm” and a “generative model”.
Search Algorithms
A search algorithm is an answer to a search problem. A search problem exists when a user desires to retrieve a bit of data or an object like a video or a song from a database. Let’s call the user’s request the query and the results of the search the response. The goal of a search algorithm is to search out that piece of data that optimally satisfies the user’s needs, i.e. provides an optimal response for the given query.
Nevertheless, there’s also a time constraint on the search problem. More often than not, we would like to receive the second-best response after 10 seconds over the best possible response after 10 hours. Due to this fact, a search algorithm should discover a response that’s qualitatively satisfactory, inside an inexpensive time.
Generative Models
A generative model is an answer to a prediction problem. Based on a set of input parameters (query), the model generates a prediction for what the optimal…