Mengjie Qian 钱梦洁

Research Associate, University of Cambridge

I am a Research Associate in the Machine Intelligence Laboratory at the University of Cambridge, working with Prof. Kate Knill and Prof. Mark Gales.

My research focuses on speech foundation models (SFMs) and large language models (LLMs), particularly their application in education and healthcare for low-resource languages.

Speech Foundation Models LLMs Spoken Language Assessment Healthcare AI
Mengjie Qian Profile Photo

Latest Highlights

Dec 2025 Awarded Cambridge Language Sciences Incubator Fund for research on Alzheimer’s Dementia detection (£7.5k).
Ongoing Lead Organiser of the Speak & Improve Challenge 2025.

Professional Experience

Research Associate Apr 2021 – Present
University of Cambridge, UK
Research Consultant May 2020 – Present
Trinity College Dublin, Ireland (Remote)
Applied Scientist Intern Dec 2018 – Mar 2019
Amazon Development Center, Cambridge, UK

Education

Ph.D. in Engineering 2016 – 2021
University of Birmingham, UK
Thesis: “Computer Analysis of Children’s Non-Native English Speech for Language Learning and Assessment”
Supervisors: Prof. Martin Russell, Dr. Peter Jančovič.
B.S. in Electronic Engineering 2010 – 2014
University of Science and Technology of China (USTC)
Supervisor: Prof. Ian McLoughlin.

Grants & Funding

Invited Talks

2025
Universal Acoustic Adversarial Attacks: From Muting to Controlling Speech Foundation and Language Models Nove 2025; University of Science and Technology of China, China
From Challenge to Solution: Speak & Improve Challenge 2025 and Advances in End-to-End Spoken Grammatical Error Correction Sep 2025; ISCA SIGML Seminar
2024
Automated Learning Teaching and Assessment: Spoken Language Processing Technology Project June 2024; Second Language Acquisition with Speech and Language Technologies (SLASaLT) Workshop, University of Sheffield

Selected Publications

Selected from 30+ peer-reviewed papers. See Google Scholar for full list.

Speak & Improve Challenge 2025: Tasks and Baseline Systems
M. Qian, K. Knill, S. Banno, S. Tang, P. Karanasou, M. J. F. Gales, D. Nicholls
Proc. ISCA SLaTE 2025
Speak & Improve Corpus 2025: an L2 English Speech Corpus for Language Assessment and Feedback
K. Knill, D. Nicholls, M. J. F. Gales, M. Qian, P. Stroinski
arXiv preprint arXiv:2412.11986, 2025
Can generative large language models perform asr error correction?
R. Ma*, M. Qian*, M. Gales, K. Knill
IEEE Transactions on Audio, Speech, and Language Processing (TASLP), 2025
Universal Acoustic Adversarial Attacks for Flexible Control of Speech-LLMs
R. Ma, M. Qian, V. Raina, M. Gales, K. Knill
EMNLP (Findings), 2025
Scaling and Prompting for Improved End-to-End Spoken Grammatical Error Correction
M. Qian, R. Ma, S. Bannò, K. Knill, M. Gales
Interspeech 2025
Learn and Don’t Forget: Adding a New Language to ASR Foundation Models
M. Qian, S. Tang, R. Ma, K. Knill, M. J. Gales
Interspeech 2024
Zero-shot Audio Topic Reranking using Large Language Models
M. Qian, R. Ma, A. Liusie, E. Loweimi, K. M. Knill, M. Gales
IEEE Spoken Language Technology (SLT) Workshop, 2025