Scholar
Jiachen Lian
Google Scholar ID: E1CsZLcAAAAJ
UC Berkeley
Citations & Impact
Citations
434
H-index
13
i10-index
14
Publications
20
Co-authors
16
list available
Publications
Resume (English only)
Academic Achievements
- Developed the first state-of-the-art speech dysfluency transcriber (UDM/SSDM), adopted by California public schools starting in 2025 to screen 1 million children
- Co-first author of 'Automated Lexical Dysfluency Analysis to Differentiate Primary Progressive Aphasia Variants', accepted as an oral presentation at AAIC 2025
- Published 'Automatic Detection of Articulatory-Based Disfluencies in Primary Progressive Aphasia' in IEEE JSTSP 2025
- Paper 'SSDM: Scalable Speech Dysfluency Modeling' accepted at NeurIPS 2024
- Recipient of the NeurIPS Scholar Award
Research Experience
- Conducts research at BAIR on human-centered, strong supervised learning beyond scaling limits
- Develops computational models to infer cognitive states from verbal dysfluencies, creating voice-based biomarkers for speech and language disorders
- Designs condition-specific AI systems for early detection, risk prediction, and therapeutic support in speech disorders
- Builds a unified platform for large-scale language function evaluation across clinical and educational settings, with HCI-driven interpretable interfaces
- Serves as a Visiting Researcher at Meta Superintelligence Lab
Background
- PhD Candidate in EECS at UC Berkeley
- Affiliated with Berkeley Artificial Intelligence Research (BAIR)
- Focuses on research questions with long-term significance, especially human-centered AI systems
- Dedicated to improving screening, diagnosis, and intervention for dyslexia and aphasia through speech AI
- Closely collaborates with Prof. Maria Luisa Gorno Tempini on clinical and educational applications of language disorder technologies
- Visiting Researcher at Meta Superintelligence Lab
