Research
I am currently a postdoctoral researcher at LS2N in Nantes, focusing on the problem of bandwidth extension, and more generally on audio synthesis, using deep learning models.
I did my PhD research on the problem of speaker localization and counting with deep neural networks.
My other research interests include music information retrieval, music generation and sound synthesis.
Publications
2023
P.-.A Grumiaux, M. Lagrange, Efficient bandwidth extension of musical signals using a differentiable harmonic plus noise model, arXiv:2311.07363, 2023.
2022
P.-.A Grumiaux, S. Kitic, L. Girin, A. Guérin, A survey of sound source localization with deep learning methods, The Journal of the Acoustical Society of America, vol. 152, no 1, p. 107-151, 2022.
2021
P.-.A Grumiaux, S. Kitic, P. Srivastava, L. Girin, A. Guérin, SALADnet: Self-Attentive multisource Localization in the Ambisonics Domain, Workshop on Applications of Signal Processing to Audio and Acoustics, New Platz, NY, 2021.
P.-.A Grumiaux, S. Kitic, L. Girin, A. Guérin, Improved feature extraction for CRNN-based multiple sound source localization, European Signal Processing Conference, Dublin, Ireland, 2021.
2020
P.-.A Grumiaux, S. Kitic, L. Girin, A. Guérin, Multichannel CRNN for speaker counting: an analysis of performance, Forum Acusticum, Lyon, France, 2020.
P.-.A Grumiaux, S. Kitic, L. Girin, A. Guérin, High-resolution speaker counting in reverberant rooms using CRNN with Ambisonics features, European Signal Processing Conference, Amsterdam, Netherlands, 2020.
PhD manuscript
2018
2021
P.-.A Grumiaux, Deep learning for speaker counting and localization with Ambisonics signals, Université Grenoble Alpes, supervised by Laurent Girin, Srdan Kitic and Alexandre Guérin.
Internship reports
2018
P.-.A Grumiaux, Automatic drums transcription using neural networks, IRCAM, supervised by Geoffroy Peeters, Master ATIAM 2017-2018.
2017
P.-.A Grumiaux, Lyrics-to-audio alignment, Audionamix, Ecole Centrale de Lille 2016-2017 (3nd and last year).
Confidential document