Zobrazeno 1 - 10
of 480
pro vyhledávání: '"Escalante, B"'
Multi-frame algorithms for single-channel speech enhancement are able to take advantage from short-time correlations within the speech signal. Deep Filtering (DF) was proposed to directly estimate a complex filter in frequency domain to take advantag
Externí odkaz:
http://arxiv.org/abs/2305.08227
Multi-frame algorithms for single-channel speech enhancement are able to take advantage from short-time correlations within the speech signal. Deep filtering (DF) recently demonstrated its capabilities for low-latency scenarios like hearing aids with
Externí odkaz:
http://arxiv.org/abs/2305.08225
Deep learning-based speech enhancement has seen huge improvements and recently also expanded to full band audio (48 kHz). However, many approaches have a rather high computational complexity and require big temporal buffers for real time usage e.g. d
Externí odkaz:
http://arxiv.org/abs/2205.05474
Complex-valued processing has brought deep learning-based speech enhancement and signal extraction to a new level. Typically, the process is based on a time-frequency (TF) mask which is applied to a noisy spectrogram, while complex masks (CM) are usu
Externí odkaz:
http://arxiv.org/abs/2110.05588
Complex-valued processing brought deep learning-based speech enhancement and signal extraction to a new level. Typically, the noise reduction process is based on a time-frequency (TF) mask which is applied to a noisy spectrogram. Complex masks (CM) u
Externí odkaz:
http://arxiv.org/abs/2006.13077
Autor:
Schröter, Hendrik, Rosenkranz, Tobias, Escalante-B., Alberto N., Zobel, Pascal, Maier, Andreas
Deep-learning based noise reduction algorithms have proven their success especially for non-stationary noises, which makes it desirable to also use them for embedded devices like hearing aids (HAs). This, however, is currently not possible with state
Externí odkaz:
http://arxiv.org/abs/2006.13067
Publikováno v:
In Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM (2019) pages 691-698
In this paper, we propose a new experimental protocol and use it to benchmark the data efficiency --- performance as a function of training set size --- of two deep learning algorithms, convolutional neural networks (CNNs) and hierarchical informatio
Externí odkaz:
http://arxiv.org/abs/1907.02549
Autor:
De-Escalante, B., Chara-Cervantes, J., Pérez-Conesa, M., Rascón, J., Pallarés, L., Perez-Guerrero, P., De-la-Red, G., Calvo, E., Soler, C., Peral-Gutiérrez, E., Gómez-Cerezo, J.F., Rodríguez-Fernández, S., Pinilla, B., Toledo-Samaniego, N., Gato, A., Chamorro, A.J., Morcillo, C., Ojeda, I., Vives, M.J., de-Miguel, B., Penadés, M., De-Vicente, M., Brito-Zerón, Pilar, Pérez-Alvarez, Roberto, Feijoo-Massó, Carles, Gracia-Tello, Borja, González-García, Andres, Gómez-de-la-Torre, Ricardo, Alguacil, Ana, López-Dupla, Miguel, Robles, Angel, Garcia-Morillo, Salvador, Bonet, Mariona, Cruz-Caparrós, Gracia, Fonseca-Aizpuru, Eva, Akasbi, Miriam, Callejas, Jose Luis, de Miguel-Campo, Borja, Pérez-de-Lis, Marta, Ramos-Casals, Manuel
Publikováno v:
In Joint Bone Spine December 2021 88(6)
Slow feature analysis (SFA) is an unsupervised-learning algorithm that extracts slowly varying features from a multi-dimensional time series. A supervised extension to SFA for classification and regression is graph-based SFA (GSFA). GSFA is based on
Externí odkaz:
http://arxiv.org/abs/1601.03945
Slow feature analysis (SFA) is an unsupervised learning algorithm that extracts slowly varying features from a time series. Graph-based SFA (GSFA) is a supervised extension that can solve regression problems if followed by a post-processing regressio
Externí odkaz:
http://arxiv.org/abs/1509.08329