Scholar
Prem Seetharaman
Google Scholar ID: XHD-48cAAAAJ
Sr. Research Scientist, Adobe Research
Citations & Impact
Citations
2,285
H-index
21
i10-index
31
Publications
20
Co-authors
67
list available
Publications
Resume (English only)
Academic Achievements
- Released VampNet (ISMIR 2023), a fast unconditional music generation model; developed the Descript Audio Codec (NeurIPS 2023), a powerful neural audio codec capable of compressing audio 90x with minimal quality loss; involved in the development of Wav2CLIP (ICASSP) and CARGAN (ICLR).
Research Experience
- Works as a Senior Research Scientist at Adobe Research in the Audio AI Lab. Previously worked at Descript on audio enhancement and generation projects.
Education
- Received his PhD in 2019 from Northwestern University, advised by Bryan Pardo.
Background
- A Senior Research Scientist at Adobe Research, focusing on generative models for all types of audio, such as everyday sounds, music, and speech. Aims to lower the barrier to entry for content creation.
Miscellany
- Shares posts on topics like computer audition, speech separation, and audio source separation on his personal blog.
