Aryan Vijay Bhosale
I am Aryan Bhosale, a MSCS graduate student in the Department of Computer Science at the University of Maryland (UMD), College Park, where I work with Prof. Dinesh Manocha (GAMMA Lab) and Prof. Ramani Duraiswami (PIRL Lab). My research interests lie in multimodal generative models, full-duplex voice agents, and music generation models.
I am an Intern at Centific AI Research where I work on full-duplex voice agents and unified audio models. I also collaborate with the Advanced Technology Group at Dolby Laboratories on building multimodal music generation models.
I completed my undergraduate studies at IIT Bombay advised by Prof. Preethi Jyothi and Prof. Biplab Banerjee. In the past, I’ve had the chance to work as a research intern at Brown University advised by Prof. Miguel Bessa and at American Express.
Please find the latest version of my Résumé/CV here.
Updates
| Oct 2026 | DuplexWorld was accepted at NeurIPS (RTCA) 2026 . |
|---|---|
| Sep 2026 | Exploring Textual Guidance for Video-to-Music Generation Models was accepted at NeurIPS 2026 Creative AI Track . |
| Aug 2026 |  VIBE was accepted at EMNLP (Findings) 2026 . |
| Aug 2026 |  Released DuplexWorld, a benchmark for speech-to-speech voice agents . |
| Jul 2026 |  Released TORUS, a self-coherence benchmark for unified audio models . |
| Jun 2026 | Joined Centific AI Research as a Research Intern. |
| Apr 2026 |  Released Video-Robin, a video-to-music generation model . |
| Sep 2025 | Started collaborating with Dolby Laboratories as a Research Collaborator. |
| Sep 2025 | Joined as a MS CS graduate student at UMD College Park. |
| May 2025 | Graduated B. Tech (with Hons.) from IIT Bombay with dual minor degrees in Artificial Intelligence and Computer Science. |
| May 2023 | Joined the Model Risk and Management Group as a Data Science Intern at American Express. |
| May 2023 | Joined Bessa Research Group as a Research Intern at Brown University. |