Scholar
Sergio Burdisso
Google Scholar ID: XOD8lrAAAAAJ
Researcher, Idiap Research Institute
Citations & Impact
Citations
503
H-index
10
i10-index
10
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Published and reviewed papers for top-tier NLP and speech venues such as ACL, EMNLP, INTERSPEECH, etc.; winner of the ICC 2024 AI Hackathon; creator of the open-source PySS3 library.
Research Experience
- Currently a Research Associate at Idiap's Speech & Audio Processing group; contributed as a senior researcher to the Automatic design of conversational models and synthetic dialog generation with LLMs at Johns Hopkins University's JSALT 2023 & 2025.
Education
- PhD, Universidad Nacional de San Luis (UNSL), Argentina. Developed the SS3 classification model during his PhD, achieving the best results in the CLEF eRisk challenge for three consecutive years (2019-2021).
Background
- Research interests: speech recognition and natural language understanding. Background: Working at the Idiap Research Institute located in a lovely valley in the heart of the Swiss Alps, focusing on coupling speech encoders with large language models (LLMs) and advancing task-oriented dialogue modeling and conversational structure induction.
Miscellany
- Nationality: Argentinian. Has been coding since childhood, with over 20 years of experience. Enjoys implementing various technologies from scratch, including communication protocols based on their RFCs (e.g., HTTP, TCP, UDP, WebSocket). Also created a C-inspired programming language for beginners and developed a library to generate valid Windows .exe (PE) files from raw x86 machine code.
Co-authors