Research interests

• Speech quality metrics
• Speech enhancement
• Unconditional audio generation
• Diffusion models
• Audio-visual speech processing
• Self-supervised learning

Publications

1. Danilo de Oliveira, Tal Peer, Timo Gerkmann, "Too Good to Be True: A Study on Modern Automatic Speech Recognition for the Evaluation of Speech Enhancement", ISCA Interspeech, Sydney, Australia, 2026. [arxiv]
2. Danilo de Oliveira, Tal Peer, Jonas Rochdi, Timo Gerkmann, "Are These Even Words? Quantifying the Gibberishness of Generative Speech Models", ICASSP, Barcelona, Spain, 2026. [doi][arxiv][code][project page]
3. Julius Richter, Danilo de Oliveira, Timo Gerkmann, "Do We Need EMA for Diffusion-Based Speech Enhancement? Toward A Magnitude-Preserving Network Architecture", ICASSP, Barcelona, Spain, 2026. [doi][arxiv][code]
4. Navin Raj Prabhu, Danilo De Oliveira, Nale Lehmann-Willenbrock, Timo Gerkmann, "Enhancing In-the-Wild Speech Emotion Conversion with Resynthesis-based Duration Modeling", ASRU, Honolulu, Hawaii, 2025. [doi][arxiv][project page]
5. Danilo de Oliveira, Julius Richter, Tal Peer, Timo Gerkmann, "Demonstration of LipDiffuser: Lip-to-Speech Generation with Conditional Diffusion Models Running on a Portable Device", WASPAA, Lake Tahoe, USA, 2025. [pdf]
6. Danilo de Oliveira, Julius Richter, Tal Peer, Timo Gerkmann, "LipDiffuser: Lip-to-Speech Generation with Conditional Diffusion Models", Preprint, 2025. [arxiv][project page]
7. Julius Richter, Danilo de Oliveira, Timo Gerkmann, "Investigating Training Objectives for Generative Speech Enhancement", ICASSP, Hyderabad, India, 2025. [doi][arxiv][code]
8. Danilo de Oliveira, Julius Richter, Jean-Marie Lemercier, Simon Welker, Timo Gerkmann, "Non-intrusive Speech Quality Assessment with Diffusion Models Trained on Clean Speech", ISCA Interspeech, Rotterdam, 2025. [doi][arxiv][code]
9. Danilo de Oliveira, Eric Grinstein, Patrick A. Naylor, Timo Gerkmann, "LASER: Language-Queried Speech Enhancer", IWAENC, Aalborg, Denmark, 2024. [doi]
10. Danilo de Oliveira, Simon Welker, Julius Richter, Timo Gerkmann, "The PESQetarian: On the Relevance of Goodhart's Law for Speech Enhancement", ISCA Interspeech, Kos, Greece, 2024. [doi][arxiv][project page]
11. Danilo de Oliveira, Timo Gerkmann, "Distilling HuBERT with LSTMs via Decoupled Knowledge Distillation", ICASSP, Seoul, Korea, 2024. [doi][arxiv] 
12. Danilo de Oliveira, Julius Richter, Jean-Marie Lemercier, Tal Peer, Timo Gerkmann, "On the Behavior of Intrusive and Non-intrusive Speech Enhancement Metrics in Predictive and Generative Settings", ITG Conference on Speech Communication, Aachen, Germany, 2023. [doi][arxiv]
13. Danilo de Oliveira, Navin Raj Prabhu, Timo Gerkmann, "Leveraging Semantic Information for Efficient Self-Supervised Emotion Recognition with Audio-Textual Distilled Models", ISCA Interspeech, Dublin, Ireland, 2023. [doi][arxiv]
14. Danilo de Oliveira, Tal Peer, Timo Gerkmann, "Efficient Transformer-based Speech Enhancement Using Long Frames and STFT Magnitudes", ISCA Interspeech, Incheon, Korea, 2022. [doi][arxiv] [audio]