Text-to-Audio GenAI: Opportunities and Challenges in Music Production and Audio Content Generation

Francesca Ronchini (Politecnico di Milano) joins us to discuss her PhD research.

Generative Artificial Intelligence (GenAI) has emerged as a significant breakthrough in recent technological history, with the potential to shape artistic practices in the coming years. These models have been successfully applied to various content types, including audio, video, and image generation. Initially, GenAI models were primarily used by machine learning and computer science professionals, with limited accessibility to the general public. However, the advent of text-based generative models, which create content based on textual descriptions, has changed this dynamic. In this talk, we will focus on text-based models for generating music and audio content, exploring the opportunities and challenges of incorporating these models into music production practices. Additionally, we will present the results of a human-centered Music Information Retrieval (MIR) study aimed at understanding how these models can be practically integrated into the artistic practices of musicians and music professionals.

NOTE: this event is part of the DL4MIR workshop series (ccrma-mir.github.io); guest speaker talks are open to the broader CCRMA community.