Scholar
Thomas Thebaud
Google Scholar ID: 5wgj-vYAAAAJ
Assistant Research Scientist, ECE Dept., Johns Hopkins University, Baltimore
Adversarial and Backdoor attacks
Speech Emotion Recognition
Audio LLMs
Speaker Characterisation
Follow
Google Scholar
↗
Citations & Impact
All-time
Citations
305
H-index
8
i10-index
7
Publications
20
Co-authors
13
list available
Contact
No contact links provided.
Publications
18 items
DiT-Flow: Speech Enhancement Robust to Multiple Distortions based on Flow Matching in Latent Space and Diffusion Transformers
2026
Cited
0
Speaker Verification with Speech-Aware LLMs: Evaluation and Augmentation
2026
Cited
0
Can LLMs Help Localize Fake Words in Partially Fake Speech?
2026
Cited
0
Reconstruct! Don't Encode: Self-Supervised Representation Reconstruction Loss for High-Intelligibility and Low-Latency Streaming Neural Audio Codec
2026
Cited
0
Spoken DialogSum: An Emotion-Rich Conversational Dataset for Spoken Dialogue Summarization
2025
Cited
0
refess-qi: reference-free evaluation for speech separation with joint quality and intelligibility scoring
2025
Cited
0
Scaling Multi-Talker ASR with Speaker-Agnostic Activity Streams
2025
Cited
0
Backdoor Attacks Against Speech Language Models
2025
Cited
0
Load more
Resume (English only)
Co-authors
13 total
Najim Dehak
Associate Professor at ECE department, Johns Hopkins University.
Co-author 2
Jesús Villalba
Johns Hopkins University
Co-author 4
Ankur Butala MD
Johns Hopkins University School of Medicine
Co-author 6
Anthony Larcher
Professor Le Mans Université
Gaël Le Lan
Meta Reality Labs
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up