Xiaoyu Lin

Xiaoyu is a PhD student co-advised with Laurent Girin. Xiaoyu investigates the use of complex deep generative models for multi-modal (audio-visual) processing. She applied these models to speech enhancement [1, 2]. Xiaoyu has also worked in mixtures of dynamical variational autoencoders, applied to source separation with audio and video data [3].

References

1. X. Lin, S. Leglaive, L. Girin, and X. Alameda-Pineda, “Speech Modeling with a Hierarchical Transformer Dynamical VAE,” in IEEE International Conference on Audio, Speech and Signal Processing, 2023. [ bib | pdf ]
   

   @inproceedings{Lin-ICASSP-2023,
     title={Speech Modeling with a Hierarchical Transformer Dynamical {VAE}},
     author={Xiaoyu Lin and Simon Leglaive and Laurent Girin and Xavier Alameda-Pineda},
     booktitle={IEEE International Conference on Audio, Speech and Signal Processing},
     year={2023},
     doi={10.1109/ICASSP49357.2023.10096751},
     pdf={http://xavirema.eu/wp-content/papercite-data/pdf/Lin-ICASSP-2023.pdf}
   }

2. X. Lin, S. Leglaive, L. Girin, and X. Alameda-Pineda, “Unsupervised speech enhancement with deep dynamical generative speech and noise models,” in Interspeech, 2023, pp. 5102-5106. [ bib | pdf | arxiv ]
   

   @inproceedings{Lin-Interspeech-2023,
     title={Unsupervised speech enhancement with deep dynamical generative speech and noise models},
     author={Xiaoyu Lin and Simon Leglaive and Laurent Girin and Xavier Alameda-Pineda},
     booktitle={Interspeech},
     year={2023},
     pages={5102-5106},
     doi={10.21437/Interspeech.2023-232},
     arxiv={https://arxiv.org/abs/2306.07820},
     pdf={http://xavirema.eu/wp-content/papercite-data/pdf/Lin-Interspeech-2023.pdf}
   }

3. X. Lin, L. Girin, and X. Alameda-Pineda, “Mixture of Dynamical Variational Autoencoders for Multi-Source Trajectory Modeling and Separation,” Transactions on Machine Learning Research, 2023. [ bib | pdf | code | arxiv ]
   

   @article{Lin-TMLR-2023,
     title={Mixture of Dynamical Variational Autoencoders for Multi-Source Trajectory Modeling and Separation},
     author={Xiaoyu Lin and Laurent Girin and Xavier Alameda-Pineda},
     year={2023},
     journal={Transactions on Machine Learning Research},
     arxiv={https://arxiv.org/abs/2312.04167},
     code={https://github.com/linxiaoyu1/MixDVAE},
     note={\url{https://openreview.net/forum?id=sbkZKBVC31}},
     pdf={http://xavirema.eu/wp-content/papercite-data/pdf/Lin-TMLR-2023.pdf}
   }