Senior Saanvi Bhargava recently presented at the International Society for Music Information Retrieval’s (ISMIR) annual conference, held this week in San Francisco. Her research, which can be viewed at the ISMIR website, delved into ways that generative models can be trained to emulate vocal styles in different genres of music. In her previous research on music technology, Bhargava investigated machine learning models that could provide accurate feedback to help singers improve their technique, which earned her one of the top five spots in The Harvard Undergraduate Research Journal’s (THURJ) Research Competition.