I am a doctoral researcher in data-driven intelligent systems at the Division of Speech, Music and Hearing (TMH) supervised by Gustav Eje Henter and Jonas Beskow. My thesis revolves around efficient and probabilistic speech synthesis. My goal is to democratize speech synthesis, ensuring accessibility for the general public rather than limiting it to large corporations. Overall, my research interests align with generative modelling, speech science (synthesis/recognition), natural language processing, and probabilistic machine learning. As a doctoral student in the field of machine learning, I am deeply passionate about the intersection of probabilistic modelling and speech synthesis. With a strong background in both mathematics and computer science, I am uniquely equipped to tackle the complex challenges that lie at the forefront of this exciting and rapidly-evolving field. My programming skills have been further honed through a range of academic and industry projects, in which I have had the opportunity to work on a variety of machine learning tasks, including the development of custom algorithms, the implementation of existing models, and the integration of these methods into production systems. My research is supported by the
Wallenberg AI, Autonomous Systems and Software Program (WASP)
Open Source dev
In addition to my programming skills, I am also an active member of the open-source community, with a number of projects hosted on GitHub. Through my work on these projects, I have had the opportunity to collaborate with a diverse group of developers and researchers, further deepening my understanding of the field and refining my skills as a researcher.
I did my masters from ITMO University, Saint Petersburg Russia and a research project at the Text-to-Knowledge lab at Ghent University Belgium. Before my masters I worked as an ETL/Java developer in Industry.
In my free time, I write at shivammehta.me (I am working on revamping this to a new platform)