Doctoral Researcher in Speech and Language Technology
Core
Developing methods to detect the origin and authenticity of AI-generated audio, including deepfake detection and content-based watermarking.
Role type
Doctoral Researcher (Speech and Language Technology)
Builds
Detection methods, datasets, and open-source software for trustworthy generative AI
Domain
Speech and audio AI, deepfake detection, watermarking
Deliverable
research
Required skills
Python programming, deep learning, signal processing, speech/audio processing, experimental design, statistical machine learning
Preferred skills
Experience with generative audio models, adversarial attack evaluation, reproducible research practices
Technologies
Python, deep learning frameworks, high-performance computing clusters (CPU/GPU)
Responsibilities
Develop detection methods for generated audio content, create diverse speech/audio datasets, adapt detector baselines for multi-objective detection, study robustness against signal-processing and adversarial attacks, publish research in leading venues
Seniority
Early-career researcher (PhD candidate)