Scholar
Tal Remez
Google Scholar ID: XqHYn7EAAAAJ
AI Researcher
Citations & Impact
Citations
15,050
H-index
23
i10-index
26
Publications
20
Co-authors
34
list available
Contact
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- Published multiple papers on topics such as discrete flow matching, improved LLM code-generation, simple and controllable music generation, textually pretrained speech language models, self-supervised speech resynthesis with visual input, visually-driven prosody for text-to-speech, on-screen sound separation, robust direct speech-to-speech translation, unsupervised audio-visual separation, image segmentation, image denoising, deep shape correspondence, point cloud sparse coding, and more.
Research Experience
- Worked at Facebook AI Research (FAIR) and Google Research, leading projects that pushed the boundaries of AI, including on-device audio-visual speech separation and advancements in LLMs.
Education
- No specific education background information provided.
Background
- An AI and machine learning researcher with a PhD. Research interests include large language models (LLMs), optimization, visual perception, computational photography, and audio-visual methods for speech enhancement. Focuses on applications of LLMs in audio and music generation, as well as text and code generation. Recently explored techniques like flow matching and diffusion in latent text embeddings.
Miscellany
- No personal interests or other related information provided.
